Research Curation Daemon

Research Curation Daemon

By Billy GlennTechnology
Download on the App Store

Research Curation Daemon episodes

  • Episode 060: You Can't Tune a Product: How a Banned Narrative Kept Shipping — and Why We Took the Pen
    You Can't Tune a Product — Show Notes

    You can tune a piano — the strings obey the wrench. You cannot tune a generative product, because its behavior is a trained prior, not a setting. This special edition tells the first-person story of the RCD pipeline itself: months of prompt detuning that failed to stop NotebookLM's "collision" dramaturgy, the forensic that proved it, and the pivot to verify-then-author — a Whisper-plus-LLM fidelity gate, then taking the pen entirely with verbatim two-host TTS.

    In this episode
    • The seduction: why NotebookLM's two-host delivery — engineered down to deliberately injected disfluencies — set the bar the pipeline is now trying to clear on its own.
    • The tell: prompts banned the "collision course" frame by name; episode 55 opened on it anyway, and episode 57 re-declared it as the show's "mission" two weeks after the fidelity gate shipped.
    • The honesty beat: the AWS single-AZ thermal event's scope drift began in-house — scout escalated, synthesis inverted the multi-AZ lesson — before NotebookLM dramatized it. The raw research was right; the "28-Hour Meltdown" analysis — an independent builder's piece on AWS's community platform, not AWS-authored — complicates the tale further. (Errata: the first cut of this episode misattributed that piece as "AWS's own" — the corrected audio confesses and explains; attribution is now a checked axis in the fidelity gate.)
    • The mechanism: reverse-engineering of NotebookLM's fixed system prompt, plus instruction-hierarchy and narrative-bias research (AAAI, IHEval, NeurIPS, EMNLP) showing why prohibitions lose to trained priors.
    • The steelman: a real AI-datacenter-load story exists under the echo-frame (IEA, LBNL, Belfer Center, Brookings) — the defect is the unearned weekly crisis thesis, not a fake topic.
    • The pivot and its critics: the STT round-trip fidelity gate and litmus library, versus the LLM-as-judge reliability literature — who gates the gate?
    • Taking the pen: the v3-vs-v2 expressiveness/fidelity trade-off, the accent-coupling disqualifier, TTS economics, Gemini's unbundled Podcasts API — and the retention data suggesting the dramaturgy was the engagement engine all along.
    • The anti-smugness beat: the gate's newest litmus entry was caught on the authored path. The Editor audits everyone, including the editor.
      • Sources & References
      Primary / originating sources (operator-provided — ground zero)
      • CHANGELOG.md (operator-provided local document)
      • done/REPORT-2026-06-21-aws-story-forensics.md (operator-provided local document)
      • specs/SPEC-2026-06-21-render-layer-fidelity.md (operator-provided local document)
      • tasks/TASK-2026-07-03-detune-collision-narrative.md (operator-provided local document)
      • outputs/bakeoff/2026-07-05/fidelity_report_notebooklm_ep057.md (operator-provided local document)
      • config/fidelity_litmus.md (operator-provided local document)
        • The product & the mechanism
        • Nicole Hennig — Reverse-Engineering NotebookLM's Audio Overview System Prompt
        • Google NotebookLM Help — Audio Overview customization limits
        • Google Cloud — Standalone NotebookLM Enterprise Podcasts API
        • AAAI 2025 — System/user prompt separation does not produce reliable hierarchy
        • Amazon Science — IHEval: evaluating models on the instruction hierarchy
        • OpenAI — The instruction hierarchy challenge
        • NeurIPS 2025 — Reasoning can degrade instruction following
        • EMNLP 2024 — GPT-4's systematic narrative-structure bias ("Man in Hole" arcs)
          • The AWS case study
          • Network World — AWS US-EAST-1 outage after data center thermal event
          • The Register — EC2 impairment in us-east-1
          • builder.aws.com — "The 28-Hour Meltdown" (independent builder analysis on AWS's community platform)
          • Akamai — When the cloud breaks: the multi-AZ ceiling
          • AWS Post-Mortem Message 41926 — the S3 capacity-removal typo
          • Tech Policy Press — cloud concentration as a governance problem
            • The grid steelman
            • IEA — Energy and AI: energy demand from AI
            • Harvard Belfer Center — AI data centers and the U.S. electric grid (LBNL synthesis; July 2024 Northern Virginia event)
            • Goldman Sachs — U.S. data center power demand projected to double by 2027
            • Brookings — the constraint is a planning gap, not a generation shortfall
              • Verification & the gate's critics
              • ACM 2025 — LLM-as-judge limits in expert domains
              • arXiv 2506.13639 — fine-tuned LLM evaluators below random on fairness tasks
              • Center for AI Policy — LLM judges are unreliable
              • Deepgram — Whisper large-v3 real-world benchmarking (competing vendor; weigh interest)
                • The render frontier & economics
                • Google Blog — Gemini 3.1 Flash TTS
                • Google AI — speech generation developer documentation
                • The Atlantic — ElevenLabs and the state of voice cloning
                • ElevenLabs — Professional Voice Cloning documentation
                • ElevenLabs — API pricing
                • Flexprice — ElevenLabs pricing breakdown
                • SQ Magazine — podcast completion statistics
                • Luciano Banchero — the "synthetic blandness" prediction
                  • Voice provenance & law
                  • Congress.gov — S.4591, NO FAKES Act (cleared Senate Judiciary June 24, 2026)
                  • EFF — the revised NO FAKES Act critique
                  • California Governor's Office — AB 2602 / AB 1836 digital likeness laws
                  • FTC — approaches to AI-enabled voice cloning (watermarking under evaluation)

                    • Have questions about this episode? Reach out.

                    26 min
                  • Episode 059: Google Rebrands Vertex AI Around Agentic Workflows
                    Episode Description

                    Google Rebrands Vertex AI Around Agentic Workflows

                    Google rebranded Vertex AI as the Gemini Enterprise Agent Platform at Cloud Next, April 22–24, positioning agents rather than model endpoints as the primary building block for enterprise AI and competing directly with Microsoft Copilot Studio and Amazon Bedrock Agents. The company simultaneously expanded its model marketplace by adding Anthropic's Claude Sonnet 5 alongside its own Gemini 3.1 Pro and Flash-Lite, and overhauled data stack naming to align with open-source conventions. Also this week, NYISO formally warned that New York City will face reliability violations from 2026 through 2030, and EU AI Act enforcement powers activate August 2 with penalties reaching 35 million euros or 7 percent of global turnover. Enterprise teams should treat both the agent platform shift and the EU deadline as requiring concrete action now.

                    Curated Research Insights
                    AI Research
                    1. International AI Safety Report 2026: No Single Alignment Strategy Can Satisfy All Stakeholders — International AI Safety Report / UK DSIT — February 2026
                      2. Credibility: Tier 2
                      Link: https://internationalaisafetyreport.org/publication/international-ai-safety-report-2026
                    2. Every Major Frontier Lab Scores D or Below; Military AI Pivot Flagged — Future of Life Institute — Summer 2026 (evidence through June 3, 2026)
                      3. Credibility: Tier 3
                      Link: https://futureoflife.org/ai-safety-index-summer-2026/
                    3. AI Incidents Up 55% in 2025; Frontier Models Degrade Under Adversarial Conditions — Stanford HAI — 2026
                      4. Credibility: Tier 2
                      Link: https://hai.stanford.edu/ai-index/2026-ai-index-report/responsible-ai
                    4. Bioweapon Risk and Evaluation Loopholes in Production Models — International AI Safety Report 2026; coverage by Inside Privacy and Heise Online — February 2026
                      5. Credibility: Tier 1 primary; Tier 3 summaries
                      Link: https://internationalaisafetyreport.org/sites/default/files/2026-02/international-ai-safety-report-2026.pdf
                    5. Nature Editorial Demands Peer-Reviewed Standards and Legal Data Accountability — Nature Vol. 649, p.7 — 2026
                      6. Credibility: Tier 1
                      Link: https://www.nature.com/articles/d41586-025-04106-0
                    6. EU AI Act August 2026 Enforcement Is Real; Only 8 of 27 States Ready — European Parliament Think Tank / EC Digital Strategy — March 18, 2026; ongoing
                      7. Credibility: Tier 2
                      Link: https://epthinktank.eu/2026/03/18/enforcement-of-the-ai-act/
                    7. European Frontier AI Entrants and Beijing Model Access Restrictions — Reuters — June 17, 2026; June 25, 2026; July 7, 2026
                      8. Credibility: Tier 4
                      Link: https://www.reuters.com/
                    8. PG&E AI Deployments Quantify Wildfire Intercepts and Reliability Gains — Renewable Energy World; PG&E Newsroom; CPUC; HBS; Redding.com — 2024–2026 context
                      9. Credibility: Tier 2–4 mixed
                      Link: https://www.pge.com/en/newsroom/currents/future-of-energy/pg-e-using-machine-learning-to-build-a-smarter--more-resilient-g.html
                      Cloud Services
                      1. Google Renames Vertex AI, Bets the Company on Agentic Workflows — Constellation Research / Google Cloud Next — April 22–24, 2026
                        2. Credibility: Tier 4 / Tier 3
                        Link: https://www.constellationr.com/insights/news/google-cloud-next-2026-look-big-themes
                      2. Google Cloud Run Gets Blackwell GPUs and Always-On Agent Runtimes — Google Cloud Blog — April 2026
                        3. Credibility: Tier 3
                        Link: https://cloud.google.com/blog/products/serverless/whats-new-for-cloud-run-at-next26
                      3. AI-Optimized Storage Tier Targets GPU Training Bottlenecks — Google Cloud Blog — April 2026
                        4. Credibility: Tier 3
                        Link: https://cloud.google.com/blog/topics/inside-google-cloud/whats-new-google-cloud
                      4. Microsoft Builds Out Its Azure Agent Stack Across Multiple Release Waves — Azure Updates / Microsoft YouTube — March–June 2026
                        5. Credibility: Tier 3
                        Link: https://azure.microsoft.com/updates
                      5. AWS Bedrock AgentCore Is the First Managed Agent Platform with Detailed Public Pricing — AWS Pricing — 2026
                        6. Credibility: Tier 3
                        Link: https://aws.amazon.com/bedrock/agentcore/pricing/
                      6. Platform Engineering Has Crossed From Differentiator to Baseline Expectation — State of Platform Engineering Vol. 4 — 2026
                        7. Credibility: Tier 3
                        Link: https://platformengineering.org/blog/announcing-the-state-of-platform-engineering-vol-4
                      7. The Cloud Waste Problem Isn’t Getting Better — Cloud Security Alliance / Flexera — June 12, 2026; ongoing
                        8. Credibility: Tier 3
                        Link: https://cloudsecurityalliance.org/blog/2026/06/12/top-cloud-cost-optimization-techniques-in-2026-for-maximum-roi
                        GridTech
                        1. NYISO Formally Warns New York City Faces Reliability Violations Starting in 2026 — Utility Dive / NYISO — 2026
                          2. Credibility: Tier 4 referencing Tier 2
                          Link: https://www.utilitydive.com/news/new-york-iso-warns-of-2026-reliability-violations-in-nyc-long-island/802699/
                        2. Pew Roadmap for Mainstreaming Virtual Power Plants — Pew Charitable Trusts — April 2026
                          3. Credibility: Tier 2
                          Link: https://www.pew.org/en/research-and-analysis/reports/2026/04/distributed-energy-can-unleash-the-resilient-affordable-grid-of-the-future
                        3. FERC Order 1920 Compliance Filings Reshape Transmission Planning — Utility Dive / FERC — 2026
                          4. Credibility: Tier 4 referencing Tier 1
                          Link: https://www.utilitydive.com/news/transmission-planning-development-aseg/811197/
                        4. Open-Source ADMS Architecture Challenges Dominant Vendors — NLR / PNNL (GridAPPS-D) — date unspecified
                          5. Credibility: Tier 2
                          Link: https://www.nlr.gov/grid/advanced-distribution-management
                        5. May 2026 Outage Data: Vehicle Strikes, Tree Contact, Equipment Failure — GridProfile — May 2026
                          6. Credibility: Tier 4
                          Link: https://gridprofile.com/insights/power-outages-may-2026/
                        6. Utility OT Cloud Challenges Air-Gap Doctrine — TRC / Verdantix / Call MC — 2026
                          7. Credibility: Tier 3
                          Link: https://www.trccompanies.com/insights/8-digital-transformation-trends-redefining-the-utility-market-in-2026/
                        7. Tax Credit Phase-Out Drives 2026 DER Surge, With a Cliff to Follow — BDO / IEA / DOE — 2025–2026
                          8. Credibility: Tier 2–3 mixed
                          Link: https://www.bdo.com/insights/industries/natural-resources/2026-natural-resources-energy-predictions
                          Security
                          1. Iran-Linked Hackers Target U.S. Grid and Water System PLCs — Utility Dive / NERC — July 2026 reporting
                            2. Credibility: Tier 4
                            Link: https://www.utilitydive.com/news/nerc-cisa-iran-war-cyber-hacking/816914/
                          2. Industrial Ransomware Evolves Toward Physical Process Disruption — CiberSafety; SANS 2026 ICS OT Survey — 2026
                            3. Credibility: Tier 4 analysis; Tier 2 survey
                            Link: https://cibersafety.com/en/Protecting-SCADA-PLCs-in-industrial-environments-by-2026/
                          3. Dragos: 119 Ransomware Groups Targeted Industrial Organizations; Control Loop Mapping — Dragos — 2026
                            4. Credibility: Tier 3
                            Link: https://www.dragos.com/blog/ot-threat-landscape-2026
                          4. 508 ICS Advisories, 2,155 CVEs; Siemens Patch Wave — Forescout / CISA / Siemens CERT — June–July 2026
                            5. Credibility: Tier 2–3 mixed
                            Link: https://www.cisa.gov/known-exploited-vulnerabilities-catalog
                          5. Poland DER Attack via Fortinet VPN — CISA Alert — February 10, 2026
                            6. Credibility: Tier 2
                            Link: https://www.cisa.gov/news-events/alerts/2026/02/10/poland-energy-sector-cyber-incident-highlights-ot-and-ics-security-gaps
                          6. Espionage Drives 71% of Utility Breaches — Verizon 2026 DBIR (via DeepStrike) — 2026
                            7. Credibility: Tier 2 primary; Tier 3 secondary
                            Link: https://www.verizon.com/business/resources/reports/dbir/
                          7. PG&E EOC Activations; NIST SP 800-82 Revision — PG&E CSR 2025; NIST — January 2026 (NIST)
                            8. Credibility: Tier 2–3 mixed
                            Link: https://csrc.nist.gov/projects/operational-technology-security
                          8. House Energy Subcommittee Advances Five Cybersecurity Bills — House Energy & Commerce Committee — 2026
                            9. Credibility: Tier 2
                            Link: https://energycommerce.house.gov/posts/energy-subcommittee-advances-five-bills-to-strengthen-american-cybersecurity

                            Have research on AI, Cloud, or Grid technology? Share your findings.

                            25 min
                          9. Episode 057: GPT-5.5 Replaces GPT-5.3 as ChatGPT's Default Model
                            Episode Description

                            GPT-5.5 Replaces GPT-5.3 as ChatGPT's Default Model

                            OpenAI made GPT-5.5 the new default in ChatGPT, rolling it out as GPT-5.5 Instant to replace GPT-5.3 Instant, with cited gains in accuracy, image understanding, STEM, and web behavior. The GPT-5.2 family was retired on June twelfth — less than three months after its own launch — a pace that signals model version is now a first-class operational dependency requiring automated evaluation and prompt migration plans. Also this week, FERC issued show-cause orders to six major grid operators and committed to national large-load interconnection rules by June 2026, and CISA's advisory coverage of OT vulnerabilities dropped to just twenty-two percent in 2025, leaving compliance programs that rely on those advisories potentially blind to most known exposures.

                            Curated Research Insights
                            AI Research

                            1. OpenAI GPT-5.5 Becomes ChatGPT's New Default Model — OpenAI — April 23–24, 2026

                              Credibility: Tier 3
                              Link: https://openai.com/index/introducing-gpt-5-5/

                            2. Open-Weight Models Now Trail Proprietary Frontier by ~3 Months — BentoML / Epoch AI / Stanford HAI / Forbes — April–May 2026

                              Credibility: Tier 3–4
                              Link: https://www.bentoml.com/blog/navigating-the-world-of-open-source-large-language-models

                            3. AI Clinical Copilot Reduces Diagnostic Errors 16% in Kenyan Primary Care Trial — OpenAI / Penda Health — 2025–2026

                              Credibility: Tier 3
                              Link: https://openai.com/index/ai-clinical-copilot-penda-health/

                            4. Stanford AI Index Documents Five-Fold Leap in Autonomous Agent Capabilities — Stanford HAI — 2026

                              Credibility: Tier 2
                              Link: https://hai.stanford.edu/ai-index/2026-ai-index-report

                            5. ISO 42001 AI Management System Certification Ecosystem Reaches Operational Maturity — CEN / NQA / Intertek / PECB — 2026

                              Credibility: Tier 2–3
                              Link: https://standards.iteh.ai/catalog/standards/cen/adc675e8-4669-4965-b4c1-c8f724832217/en-iso-iec-42001-2026

                            6. OpenAI Proposes Aviation-Style Cross-Industry AI Incident Database — OpenAI Forum — April 6, 2026

                              Credibility: Tier 3
                              Link: https://forum.openai.com/public/videos/event-replay-sam-altman-on-building-the-future-of-ai-2026-04-06

                            7. PG&E Deploys Generative AI at Diablo Canyon and Launches Transmission Sensor Demonstration — Utility Dive / PG&E — 2024–2025

                              Credibility: Tier 3–4
                              Link: https://investor.pgecorp.com/news-events/press-releases/press-release-details/2025/PGE-Powers-Ahead-on-Breakthrough-Grid-Innovation-with-Dynamic-Line-Rating-Asset-Health-Monitoring-Demonstration/default.aspx

                              Cloud Services

                              1. AWS Launches Amazon Quick, Its Answer to Microsoft Copilot — AWS Blog — 2026

                                Credibility: Tier 3
                                Link: https://aws.amazon.com/blogs/aws/top-announcements-of-the-whats-next-with-aws-2026/

                              2. AWS Graviton5 Arrives with 25% Compute Gain, Plus CloudFormation and EKS Updates — AWS Blog — 2026

                                Credibility: Tier 3
                                Link: https://aws.amazon.com/blogs/aws/

                              3. AI API Pricing in Mid-2026 — Where OpenAI and Anthropic Actually Diverge — FinOut / Vantage / Stack Spend / MorphLLM / CloudZero — 2026

                                Credibility: Tier 3–4
                                Link: https://www.finout.io/blog/openai-vs-anthropic-api-pricing-comparison

                              4. Nuclear Power Race — Hyperscalers Bet Big on Reactors for AI Data Centers — Data Center Dynamics — 2024

                                Credibility: Tier 4
                                Link: https://www.datacenterdynamics.com/en/news/new-nuclear-capacity-has-potential-to-meet-10-percent-of-data-center-demand-in-us-over-next-decade-report/

                              5. Google Patches Vertex AI Workbench Credential Theft Vulnerability — SentinelOne / NVD — February 26, 2026

                                Credibility: Tier 3
                                Link: https://nvd.nist.gov/vuln/detail/CVE-2026-2244

                              6. DORA, the EU AI Act, and the DOJ Data Security Program Converge on Cloud Compliance — Morgan Lewis / Prime Secured / TrustArc — 2025–2026

                                Credibility: Tier 3–4
                                Link: https://www.morganlewis.com/pubs/2026/03/cybersecurity-privacy-2026-enforcement-regulatory-trends

                              7. Germany's BSI Publishes Concrete Cloud Sovereignty Standard — CSO Online — April 2026

                                Credibility: Tier 4
                                Link: https://www.csoonline.com/article/4184634/sovereign-cloud-wont-fix-your-ai-risk-identity-governance-will.html

                                GridTech

                                1. FERC Issues Show-Cause Orders to Six Grid Operators — National Data Center Interconnection Rules Due by June 2026 — Utility Dive / FERC / White & Case — 2025–2026

                                  Credibility: Tier 4 reporting on Tier 1 action; Tier 3 legal analysis
                                  Link: https://www.whitecase.com/insight-alert/ferc-orders-grid-operators-promptly-revise-or-justify-interconnection-rules-data

                                2. ERCOT Queue Hits 410 GW — Texas Proposes $50K/MW Deposit to Filter Phantom Load — ERCOT / PUCT — April 1, 2026

                                  Credibility: Tier 2
                                  Link: https://www.ercot.com/files/docs/2026/04/01/ERCOT_LargeLoad_Update_April2026_B-C_-Hearing.pdf

                                3. NextEra Energy and Dominion Energy Announce $67B Merger — Data Centers Are the Strategic Driver — NextEra / Reuters / Politico / Inside Climate News — May 18–19, 2026

                                  Credibility: Tier 3; Tier 4
                                  Link: https://newsroom.nexteraenergy.com/2026-05-18-NextEra-Energy-and-Dominion-Energy-to-Combine,-Creating-the-Worlds-Largest-Regulated-Electric-Utility-Business-and-North-Americas-Premier-Energy-Infrastructure-Platform-Benefiting-Customers?l=12

                                4. Alphabet Acquires Intersect Power for $4.75B — Tech Companies Building Their Own Grid — PwC / Washington Post — February–May 2026

                                  Credibility: Tier 3–4
                                  Link: https://www.pwc.com/us/en/industries/energy-utilities-resources/library/power-utilities-deals-outlook.html

                                5. Duke Energy's Carolinas Merger Advances — Google and Walmart Are Formal Stakeholders — Duke Energy / trade coverage — January–February 2026

                                  Credibility: Tier 4
                                  Link: https://www.linkedin.com/pulse/duke-energy-hails-1-billion-savings-merger-two-carolina-9mbke

                                6. Grid-Enhancing Technologies Move from Pilots to Mandates — PJM Proposes Requiring Dynamic Line Ratings on High-Congestion Lines — Utility Dive / FERC — 2024–2026

                                  Credibility: Tier 4
                                  Link: https://www.utilitydive.com/news/states-doe-modern-grid-deployment-initiative-gets/717338/

                                7. PG&E's Data Center Pipeline Reaches 10 GW — But Flex Connect's Scalability Is Unproven — PG&E Investor Relations / CPUC / Utility Dive — 2025

                                  Credibility: Tier 2–4
                                  Link: https://investor.pgecorp.com/news-events/press-releases/press-release-details/2025/PGE-Data-Center-Demand-Pipeline-Swells-to-10-Gigawatts-with-Potential-to-Unlock-Billions-in-Benefits-for-California/default.aspx

                                8. J.P. Morgan Projects $5.8 Trillion in Global Grid Investment Through 2035 — Digital Grid Gets $700 Billion — J.P. Morgan / Bank of America Institute — 2026

                                  Credibility: Tier 3
                                  Link: https://www.jpmorgan.com/insights/sustainability/climate/grid-resilience-neglected-no-more

                                  Security

                                  1. CISA's OT Advisory Coverage Collapses to Twenty-Two Percent — Forescout — 2026

                                    Credibility: Tier 3
                                    Link: https://www.forescout.com/blog/ics-cybersecurity-in-2026-vulnerabilities-and-the-path-forward/

                                  2. Federal Agencies Jointly Release Zero Trust Guidance for OT Environments — CISA/DOE/FBI/State Dept. — April 2026

                                    Credibility: Tier 2
                                    Link: https://www.cisa.gov/topics/industrial-control-systems

                                  3. Ransomware Against Industrial Sectors Up Fifty-Eight Percent in 2025; Malware Now Targets Physical Processes — Nexus Connect / Dragos — 2026

                                    Credibility: Tier 3–4
                                    Link: https://www.dragos.com/blog/ot-threat-landscape-2026

                                  4. Routers Now Account for Majority of Dangerous IoT Vulnerabilities; Energy Sector Attacks Up Four Hundred Fifty-Nine Percent — Swif AI / Ordr / Vectra AI — 2025–2026

                                    Credibility: Tier 3–4
                                    Link: https://ordr.net/blog/iot-security-statistics

                                  5. BadBox 2.0 Pre-Infects Ten Million Devices Before Deployment — Vectra AI / Nozomi Networks — February 2026

                                    Credibility: Tier 3
                                    Link: https://www.nozominetworks.com/ot-iot-cybersecurity-trends-insights-february-2026

                                  6. Microsoft Security Copilot Moves to Agentic Architecture, Bundled in E5/E7 — Microsoft / Logicalis / AlphaBold / GDT — April–June 2026

                                    Credibility: Tier 3
                                    Link: https://www.microsoft.com/en-us/security/blog/2026/06/02/microsoft-build-2026-securing-code-agents-and-models-across-the-development-lifecycle/

                                  7. NERC CIP 2026 Enforcement: Low-Impact Assets, MFA, and Supply Chain Now in Scope — Xage / Industrial Defender / PG&E — 2025–2026

                                    Credibility: Tier 3
                                    Link: https://xage.com/blog/navigating-the-nerc-cip-recent-revisions-with-xage/

                                    Have research on AI, Cloud, or Grid technology? Share your findings.

                                    28 min
                                  8. Episode 056: FERC standardizes grid rules for AI data centers
                                    Episode Description

                                    FERC standardizes grid rules for AI data centers

                                    The Federal Energy Regulatory Commission officially standardized the rulebook for connecting massive AI data centers to the transmission grid. With thirty percent of US data centers already pivoting to islanded microgrids to bypass backlogs, this federal clarity arrives just as utilities like PG and E roll out a seven hundred ninety-one million dollar portfolio for one-minute algorithmic load control. In a synchronized push for grid flexibility, regulators also rejected restrictive single-node caps on distributed energy resources to allow for multi-asset aggregation. Developers can now confidently secure financing and structure hybrid power agreements without getting trapped in case-by-case regulatory ambiguity.

                                    Curated Research Insights
                                    AI Research
                                    1. Stanford AI Index: Top Frontier Models Have Converged — No Clear “Best Model” - Stanford HAI - 2026
                                      2. Credibility: Tier 2
                                      Link: https://hai.stanford.edu/ai-index/2026-ai-index-report/technical-performance
                                    2. Stanford IRSL: New Psychometrics-Derived Method Cuts Scaling-Law Estimation Compute by 99% - Stanford HAI - 2025
                                      3. Credibility: Tier 2
                                      Link: https://hai.stanford.edu/news/new-approach-to-scaling-laws-could-change-how-ai-models-are-trained
                                    3. AI Incidents Rose 55% in 2025; Safety Disclosure Gaps Persist - Stanford HAI / AI Incident Database - 2026
                                      4. Credibility: Tier 2 (Stanford), Tier 3 (AIID)
                                      Link: https://hai.stanford.edu/ai-index/2026-ai-index-report/responsible-ai
                                    4. Anthropic Withholds “Most Capable and Riskiest” Model; Publishes Detailed Safety Cards - Anthropic - Feb–Jun 2026
                                      5. Credibility: Tier 3 (vendor), Tier 4 (analyst commentary)
                                      Link: https://www.anthropic.com/transparency
                                    5. Benchmark Saturation: Which Evals Still Have Signal in 2026 - Stanford HAI / Community Synthesis - 2026
                                      6. Credibility: Tier 2 (Stanford), Tier 4 (LLM Stats), Tier 5 (community list)
                                      Link: https://www.reddit.com/r/LocalLLaMA/comments/1rovfbw/i_made_a_list_of_every_ai_benchmark_that_still/
                                    6. Google Launches “Gemini for Science” Suite at I/O 2026 - Google - May–Jun 2026
                                      7. Credibility: Tier 3
                                      Link: https://blog.google/innovation-and-ai/technology/research/gemini-for-science-io-2026/
                                    7. Eli Lilly Deploys Pharma’s Most Powerful AI Supercomputer; MIT Advances Protein Drug Design - Crescendo AI / Cell Reports Medicine - 2026
                                      8. Credibility: Tier 4 (aggregator), Tier 1 (journal)
                                      Link: https://www.crescendo.ai/news/ai-breakthroughs-and-latest-ai-news-2026
                                    8. PG&E’s $791M Demand Response Plan Integrates Real-Time AI Load Control - CPUC / PG&E - 2024–2027 filing, active 2026 programs
                                      9. Credibility: Tier 2 (CPUC), Tier 3 (PG&E)
                                      Link: https://docs.cpuc.ca.gov/PublishedDocs/SupDoc/A2205002/4857/474109675.pdf
                                      Cloud Services
                                      1. Kubernetes v1.33 Reaches End of Support Today — Immediate Security Risk for Lagging Clusters - Kubernetes Project / endoflife.date - June 28, 2026
                                        2. Credibility: Tier 1–2
                                        Link: https://endoflife.date/kubernetes
                                      2. Ingress NGINX Retired by SIG Network in March 2026 - Kubernetes Blog - March 30, 2026
                                        3. Credibility: Tier 1
                                        Link: https://kubernetes.io/blog/2026/03/30/kubernetes-v1-36-sneak-peek/
                                      3. Kubernetes v1.37 Feature Set Locked — August 26, 2026 Release on Track - Kubernetes Dev - 2026
                                        4. Credibility: Tier 1–2
                                        Link: https://www.kubernetes.dev/resources/release/
                                      4. OpenTelemetry Production Use Doubles to 11%, But “Universal Standard” Narrative Outpaces Reality - Elastic - 2026
                                        5. Credibility: Tier 3
                                        Link: https://www.elastic.co/blog/2026-observability-trends-generative-ai-opentelemetry
                                      5. Agentic AI Drives 5–10x Telemetry Volume Spike — Forcing Observability Cost Reckoning - Elastic / Grafana / IBM / Dynatrace - 2026
                                        6. Credibility: Tier 3
                                        Link: https://grafana.com/observability-survey/
                                      6. Vector Database Market Matures, But Benchmark Figures Warrant Skepticism - Salt Techno AI / Actian - 2026
                                        7. Credibility: Tier 5 (benchmark), Tier 3 (Actian)
                                        Link: https://www.actian.com/blog/databases/how-to-evaluate-vector-databases-in-2026/
                                      7. Platform Engineering Crosses Maturity Threshold in 2026 — With Caveats on Source Quality - Adaptavist / DZone / Recast - 2026
                                        8. Credibility: Tier 3–4
                                        Link: https://www.adaptavist.com/blog/devops-2026-predictions-what-the-experts-really-think
                                        GridTech
                                        1. FERC Delivers Comprehensive Large-Load Interconnection Rules at June 18 Meeting - FERC / CAISO / White & Case - June 18, 2026
                                          2. Credibility: Tier 2
                                          Link: https://www.ferc.gov/news-events/news/summaries-june-2026-commission-meeting
                                        2. FERC Rejects PJM’s Single-Node Limit on DER Aggregations - NARUC - 2023–2024 summary, active in 2026 context
                                          3. Credibility: Tier 2
                                          Link: https://pubs.naruc.org/pub/0C205FC7-C2D1-071A-B5F3-70BF27231837
                                        3. Order 2023 Reform Wave Approaches a National Synchronized Test This Fall - Zonevex / Wood Mackenzie / LBNL - June 2026
                                          4. Credibility: Tier 2–4
                                          Link: https://zonevex.com/cluster-calendar
                                        4. GM Launches Bidirectional EV Rollout, Names PG&E as Utility Partner - GM News / UC Berkeley ITS - June 9, 2026
                                          5. Credibility: Tier 3 (GM), Tier 2 (Berkeley ITS)
                                          Link: https://news.gm.com/home.detail.html/Pages/news/us/en/2026/jun/0609-bidirectional-evs.html
                                        5. Europe’s First Islanded Microgrid Data Center Opens in Dublin as U.S. Adoption Reaches Thirty Percent - CNBC / IEEE - March 11, 2026
                                          6. Credibility: Tier 4 (CNBC), Tier 1 (IEEE)
                                          Link: https://www.cnbc.com/2026/03/11/data-center-microgrid-power-ireland-ai-boom-avk-pure-dc.html
                                        6. Kairos Power Breaks Ground on Hermes Reactor While DOE Commits $800M to Two SMR Sites - ASME / DOE / EIA - 2025–2026
                                          7. Credibility: Tier 2–3
                                          Link: https://www.asme.org/topics-resources/content/what-nuclear-energy-technologies-are-actually-advancing-in-2026
                                        7. Smart Meter Industry Presses Utilities on AMI 2.0 While Deployments Hit Global Scale - Latitude Media / LADWP / WSSC - 2025–2026
                                          8. Credibility: Tier 2 (utilities), Tier 4 (trade media)
                                          Link: https://www.latitudemedia.com/news/will-utilities-deploy-next-generation-meters-to-support-a-smarter-more-flexible-grid/
                                          Security
                                          1. U.S. Executive Order Sets 2030 Hard Deadline for Federal Post-Quantum Migration - Federal reporting aggregation - June 22, 2026
                                            2. Credibility: Tier 4 (verify against Federal Register)
                                            Link: https://csrc.nist.gov/projects/post-quantum-cryptography
                                          2. EU Cyber Resilience Act Vulnerability Reporting Now Active — Smart Meter Gateways Included - CEN-CENELEC / IPC2U - Feb 23, 2026 webinar; dates into 2026–2027
                                            3. Credibility: Tier 2 (CEN-CENELEC), Tier 4 (IPC2U)
                                            Link: https://www.cencenelec.eu/media/CEN-CENELEC/Events/Webinars/2026/2026-02-23_cra-standards-unlocked-cybersecurity-requirements-for-smgw_final.pdf
                                          3. IEC 62443-4-1 Will NOT Provide CRA Presumption of Conformity - Sarah Fluchs / iTeh Standards - 2025–2026
                                            4. Credibility: Tier 3 (named expert)
                                            Link: https://standards.iteh.ai/catalog/standards/clc/6c82a95f-8ca7-4d82-af04-a2f45052de23/en-iec-62443-4-1-2018-praa-2026
                                          4. IEC 62443 Series Expands to Six Security Categories — IIoT and Third-Party Assessment Now Covered - ABHISAM - 2026
                                            5. Credibility: Tier 4
                                            Link: https://www.abhisam.com/iec-62443-latest-version-updates-in-2026/
                                          5. Itron Smart Meter Vendor Confirmed Intrusion; UK Hands Security Oversight to Industry - Cybersecurity Dive / NCSC - Apr 2024; Sep 2025
                                            6. Credibility: Tier 4 (incident), Tier 2 (policy)
                                            Link: https://www.ncsc.gov.uk/blog-post/ncsc-handing-over-baton-smart-meter-security
                                          6. EU CRA Creates Active AMI Security Mandates — Lateral Movement as Primary Systemic Risk - IoT Use Case / Telit / Smart DCC - 2026
                                            7. Credibility: Tier 3–4
                                            Link: https://www.smartdcc.co.uk/our-smart-network/protecting-data-on-the-smart-network/
                                          7. CISA Introduces POEM Framework for Critical Infrastructure Insider Threat Programs - Industrial Cyber / CDSE - 2026
                                            8. Credibility: Tier 2 (CDSE), Tier 4 (Industrial Cyber)
                                            Link: https://industrialcyber.co/cisa/cisa-introduces-poem-framework-to-strengthen-insider-threat-mitigation-across-critical-infrastructure/
                                          8. Tenable Joins OpenAI Daybreak Program; Launches AI-Powered Cloud Detection and Response - Tenable / GlobeNewswire - May–June 2026
                                            9. Credibility: Tier 3
                                            Link: https://investors.tenable.com/news-releases

                                            Have research on AI, Cloud, or Grid technology? Share your findings.

                                            21 min
                                          9. Episode 055: FERC pushes public disclosure for grid cyber violators
                                            Episode Description

                                            FERC pushes public disclosure for grid cyber violators

                                            Federal energy regulators are advancing a joint proposal to publicly name utility companies that violate grid cybersecurity standards, ending over a decade of enforcement anonymity. Under the proposed framework, violators facing statutory civil penalties of up to one million dollars per day will also contend with severe reputational exposure. While the utility industry has raised concerns about providing actionable intelligence to potential adversaries, regulators explicitly separated entity identities from sensitive technical vulnerabilities. Compliance teams must immediately prepare for a new era of grid enforcement where cyber failures carry public accountability stakes similar to data breach notifications in other critical sectors.

                                            Curated Research Insights
                                            AI Research

                                            1. NVIDIA, AMD, and Intel All Launch Next-Generation AI Platforms in 2026 - NVIDIA IR / Reuters / S&P Global / Intel Newsroom - Jan 5, 2026; May 31, 2026; Mar 2026; Jun 2026

                                              Credibility: Tier 3/4/4/3
                                              Link: https://investor.nvidia.com/news/press-release-details/2026/NVIDIA-Kicks-Off-the-Next-Generation-of-AI-With-Rubin--Six-New-Chips-One-Incredible-AI-Supercomputer/default.aspx

                                            2. Researchers Declare the Foundations of Explainable AI Are Broken - arXiv / Nature npj AI - Feb 2026; 2026

                                              Credibility: Tier 2 / Tier 1
                                              Link: https://arxiv.org/abs/2602.24176

                                            3. PG&E's AI-Driven Grid Hardening Program Hits 1,260 Miles, Cuts Costs by 20% - Plumas Sun / PG&E IR - Oct 23, 2025; Nov 2024

                                              Credibility: Tier 4 / Tier 3
                                              Link: https://plumassun.org/2025/10/23/1000-miles-of-powerlines-now-underground-says-pge/

                                            4. Transformers Aren't Being Replaced — They're Being Extended - Independent blog / Nebius / Towards AI - 2024–2025 (various)

                                              Credibility: Tier 5 / Tier 3 / Tier 4
                                              Link: https://philippdubach.com/posts/the-last-architecture-designed-by-hand/

                                            5. Applied Energy Publishes Peer-Reviewed Work on AI for Renewable Integration - Applied Energy - 2026

                                              Credibility: Tier 1
                                              Link: https://www.sciencedirect.com/science/article/pii/S0306261926003107

                                            6. Mechanistic Interpretability Pushes Beyond Language Models Into Algorithmic Reasoning - arXiv - Feb 2026

                                              Credibility: Tier 2
                                              Link: https://www.arxiv.org/list/cs.AI/2026-02?skip=3450&show=2000

                                              Cloud Services

                                              1. AWS US-East-1 Goes Dark for Twenty-Eight Hours After Thermal Event - Network World / AWS Builder Community - May 2026

                                                Credibility: Tier 4 / Tier 3
                                                Link: https://www.networkworld.com/article/4168878/aws-hit-by-us-east-1-outage-after-data-center-thermal-event.html

                                              2. Storm-2949 Shows How Stolen Credentials Cascade Into Full Cloud Compromise - Microsoft Security Blog - May 18, 2026

                                                Credibility: Tier 3
                                                Link: https://www.microsoft.com/en-us/security/blog/2026/05/18/storm-2949-turned-compromised-identity-into-cloud-wide-breach/

                                              3. Cloud Security Alliance Names Machine Identities as the Top Cloud Risk of 2026 - Cloud Security Alliance - Mar 13, 2026

                                                Credibility: Tier 2
                                                Link: https://cloudsecurityalliance.org/blog/2026/03/13/the-state-of-cloud-and-ai-security-in-2026

                                              4. Wiz and Google Announce AI-Native Security Platform at RSA - Google Cloud Blog - 2026

                                                Credibility: Tier 3
                                                Link: https://cloud.google.com/blog/products/identity-security/next26-redefining-security-for-the-ai-era-with-google-cloud-and-wiz

                                              5. Developer Toolchain Under Attack — Vercel, LiteLLM, and VS Code Among 2026 Supply Chain Targets - PKWARE - 2026

                                                Credibility: Tier 4
                                                Link: https://www.pkware.com/blog/2026-data-breaches

                                              6. PG&E Surpasses Twelve Hundred Miles of Undergrounded Lines, With Machine Learning Driving Prioritization - PG&E / Spatial Business Systems - Jun 1, 2026; undated

                                                Credibility: Tier 2 / Tier 4
                                                Link: https://www.pge.com/en/outages-and-safety/safety/community-wildfire-safety-program/system-hardening-and-undergrounding.html

                                              7. Liquid Cooling Moves From Specialty Option to Engineering Requirement for AI Data Centers - CoreSite / Data Center World / Tom's Hardware - 2025–2026

                                                Credibility: Tier 3 / Tier 4 / Tier 4
                                                Link: https://www.coresite.com/blog/data-center-outlook-2026-power-and-cooling-challenges-and-solutions-are-top-of-mind

                                              8. Microservices Aren't the Right Default — A Practitioner Counterpoint to Cloud-Native Orthodoxy - LinkedIn / Red Hat - 2026

                                                Credibility: Tier 4 / Tier 3
                                                Link: https://www.linkedin.com/pulse/cloud-native-best-practices-2026-custom-development-tabeen-meraj-ziowc

                                                GridTech

                                                1. U.S. EV Charging Network Hits 242,000 Ports — Federal NEVI Funding in Doubt - Aftermarket Matters / IECI - 2026; Jan 21, 2026

                                                  Credibility: Tier 4 / Tier 2
                                                  Link: https://www.aftermarketmatters.com/national-news/the-2026-ev-charging-station-report-state-by-state-breakdown/

                                                2. California Presses Ahead on EV Infrastructure Despite Federal Headwinds - California Energy Commission - 2025–2026

                                                  Credibility: Tier 2
                                                  Link: https://www.energy.ca.gov/programs-and-topics/programs/federal-ev-infrastructure-programs

                                                3. PG&E Files $18.9 Billion Wildfire Mitigation Plan for 2026–2028 - PG&E / RCRC - 2026

                                                  Credibility: Tier 3 / Tier 4
                                                  Link: https://www.pge.com/assets/pge/docs/outages-and-safety/outage-preparedness-and-support/pge-2026-2028-base-wmp-vol1-r0.pdf

                                                4. Wildfire Mitigation Planning Goes Multi-State — LS Power, PacifiCorp, Portland General Electric - Energy Safety / Utility Filings - Jan 28, 2026; 2026

                                                  Credibility: Tier 2 / Tier 3
                                                  Link: https://energysafety.ca.gov/news/2026/01/28/energy-safety-issues-approval-of-ls-powers-2026-2028-wildfire-mitigation-base-plan/

                                                5. NREL Research Finds Climate Change Will Systematically Shift Solar and Wind Resource Profiles — Applied Energy / NREL - 2026

                                                  Credibility: Tier 1
                                                  Link: https://research-hub.nrel.gov/en/publications/assessing-climate-change-induced-variability-in-generation-potent/

                                                6. IISS Reframes Volt Typhoon as Disruption Pre-Positioning, Not Espionage - Industrial Cyber / NJ OHSP - 2026

                                                  Credibility: Tier 4 / Tier 2
                                                  Link: https://industrialcyber.co/industrial-cyber-attacks/iiss-notes-volt-typhoons-targeting-of-us-infrastructure-signals-disruptive-intent-beyond-espionage/

                                                7. U.S. Battery Storage Market Posts Record Q1 2025, Up 32% Year Over Year - Utility Dive / SEIA - 2025

                                                  Credibility: Tier 4
                                                  Link: https://www.utilitydive.com/news/us-energy-storage-installations-hit-q1-record-up-32-year-over-year-seia/821133/

                                                  Security

                                                  1. FERC Proposes Named Disclosure for Grid CIP Violations - FERC / NERC - 2026

                                                    Credibility: Tier 2
                                                    Link: https://www.ferc.gov/news-events/news/comments-sought-fercnerc-staff-white-paper-cip-standards-notices-penalties

                                                  2. Dragos Publishes Ninth Annual OT Cybersecurity Year in Review - Dragos / SANS - 2026

                                                    Credibility: Tier 3 / Tier 2
                                                    Link: https://www.dragos.com/ot-cybersecurity-year-in-review

                                                  3. UK Government Formalizes Five-Year Energy Cyber Strategy — Names ICS Targeting as Strategic Threat - UK Government / CSIS / DOE CESER - 2026; 2025

                                                    Credibility: Tier 2 / Tier 2 / Tier 2
                                                    Link: https://www.gov.uk/government/publications/energy-sector-cyber-security-strategy/energy-sector-cyber-security-strategy

                                                  4. Salt Typhoon Continues Into 2026; Volt Typhoon Assessed as Pre-Positioned for Disruption - Trend Micro / Unit 42 / IISS / NJ Cyber - Apr 2026; 2026

                                                    Credibility: Tier 3 / Tier 3 / Tier 4 (citing Tier 2) / Tier 2
                                                    Link: https://unit42.paloaltonetworks.com/volt-typhoon-threat-brief/

                                                  5. CISA Initiative: Critical Infrastructure Must Function Under Active Attack - The Record / CFR - 2026; 2017 (context)

                                                    Credibility: Tier 4 / Tier 2
                                                    Link: https://therecord.media/cisa-initiative-aims-for-critical-infrastructure-to-operate-during-cyberattacks

                                                  6. CIRCIA Reporting Requirements and Mutual Aid Architecture Now Active for Utilities - APPA / NGA - 2026

                                                    Credibility: Tier 3 / Tier 2
                                                    Link: https://www.publicpower.org/system/files/documents/23%202026%20Issue%20Briefs_Grid%20Security_FINAL.pdf

                                                  7. ISA/IEC 62443 and Zero-Trust Frameworks — The OT Standards Tension Worth Watching - Dragos / NIST SP 800-207 - 2026; 2020 (context)

                                                    Credibility: Tier 3 / Tier 1
                                                    Link: https://www.dragos.com/blog/isa-iec-62443-concepts

                                                    Have research on AI, Cloud, or Grid technology? Share your findings.

                                                    15 min
                                                  8. Episode 054: The Bounds of the Cloned Voice
                                                    The Bounds of the Cloned Voice

                                                    In 2024, journalist Evan Ratliff cloned his own voice, wired it to a chatbot, and let it loose on the phone — and the thing that gave it away was a pause. This deep dive traces where that pause went. By 2026 the technical tells Ratliff relied on — latency, woodenness, the audible seams — have largely collapsed, which means the real boundary of the cloned voice has moved off the machine and onto us: our psychology, our laws, and the economics of who can afford to fake whom. The defense that still holds isn't a better detector. It's whether you and the person on the other end agreed on a code word in advance.

                                                    In this episode
                                                    • The 2024 baseline, from the source. Evan Ratliff's Shell Game — the three-part voice-agent stack (ElevenLabs clone + LLM "brain" + phone number), the DIY rig that off-the-shelf platforms quickly obsoleted, and the three honest tells: a robotic pause, a wooden affect, and a "world-class bullshitter" that confabulated zip codes.
                                                    • The tells fall. Native speech-to-speech models (ICLR 2026; Hume EVI 3) and sub-300ms latency engineer out the pause Ratliff exploited — while the "indistinguishable clone" claim rests on marketing language with no published benchmarks behind it.
                                                    • Believability was never in the machine. The psychology research confirms Ratliff's hunch: primed expectations, emotional arousal, and years of bad-VoIP conditioning do the attacker's work. Seventy percent of people can't reliably tell a clone from a real voice.
                                                    • The economics that make it inevitable. All-in automated vishing at two to six cents a minute; voice generation is now the cheapest part of the fraud stack. The same cost collapse is reshaping voice-actor labor — but vendors themselves concede the performance gap.
                                                    • Consent, the dead, and the law racing to catch up. AB 1836, the ELVIS Act, and the NO FAKES Act's June 2026 committee advance — against Berkeley Law's finding that no will can stop cloning if your audio is public, and Cambridge's warning about griefbots "haunting" the bereaved.
                                                    • The escalation: when the camera stops being proof. Ratliff turned his video on to prove he was human; FBI advisories now document multimodal "proof of life" fakes in virtual-kidnapping scams — and detection is measurably behind the attack.
                                                    • The contrarian beat. Clones still fail, behavioral defenses outperform technical ones, the scariest fraud-economics numbers are vendor estimates, and the headline detection benchmark was self-published by a contestant.
                                                      • Sources & References
                                                      Primary / originating sources (operator-provided — ground zero)
                                                      • https://www.shellgame.co/
                                                      • https://davidepstein.substack.com/p/attack-of-the-ai-voice-clones
                                                      • https://radiolab.org/podcast/shell-game/transcript
                                                        • Research & critique (peer-reviewed and academic)
                                                        • ICLR 2026 — Speech-to-Speech LLM (Zhao et al., OpenReview)
                                                        • Berkeley Law — "Vocal Identity Under Siege" (Lee & Sun, Oct 2025, PDF)
                                                        • Cambridge / Philosophy and Technology — Safeguards against AI "hauntings" by deadbots
                                                        • AI and Ethics — Non-addiction as a griefbot design principle (ScienceDirect, 2026)
                                                        • Computers in Human Behavior — Scam-call psychology / ScamGen (2024)
                                                        • Florida International University — Model-statement deception detection (2026)
                                                        • arXiv — AT-ADD Grand Challenge audio deepfake benchmark (Apr 2026)
                                                        • Hugging Face — Human Perception Audio Deepfake 2026 dataset
                                                        • ACM Digital Library — Digital resurrection misuse research
                                                          • Law, regulation & government
                                                          • FBI IC3 — Virtual kidnapping / AI voice cloning advisory (PSA251205, Dec 5, 2025)
                                                          • FBI IC3 2025 losses — reporting via Malwarebytes (Jun 2026)
                                                          • FTC — Preventing Harms from AI-Enabled Voice Cloning (Nov 2023)
                                                          • NO FAKES Act clears Senate Judiciary Committee (Deadline, Jun 2026)
                                                          • California AB 1836 — full text (LegiScan; operative Jan 1, 2025)
                                                          • Senate JEC — Sen. Hassan presses AI voice-cloning companies (Apr 2026)
                                                          • UN News — AI-enabled fraud as a cross-border challenge (Mar 2026)
                                                            • Industry, implementation & economics
                                                            • Hume AI — EVI 3 (native speech-to-speech)
                                                            • Inworld AI — Best speech-to-speech model comparison (latency)
                                                            • Cresta — Engineering for real-time voice agent latency
                                                            • Cerebrium — Global-scale voice agent at 500ms latency
                                                            • Deepgram — 2026 Voice AI Buyer's Guide (platform pricing)
                                                            • Vellum AI — Voice agent platforms guide
                                                            • Forasoft — Real-time voice cloning fraud-stack economics
                                                            • Resemble AI — Pricing (synthesis cost floor)
                                                            • Resemble AI — 2026 eight-system detection benchmark (vendor-self-published; see caveat)
                                                            • McAfee — AI voice scam research (3-second clone; 70% can't detect)
                                                            • Loeb & Loeb — SAG-AFTRA / Replica Studios AI voice agreement
                                                            • Forbes — Virginie Berger on SAG-AFTRA's AI voice gamble
                                                            • Respeecher — How character AI voice works (vendor concedes performance gap)
                                                            • Murf AI — Will AI replace voice actors (vendor self-assessment)
                                                            • ElevenLabs — Professional Voice Cloning documentation
                                                            • DeepIDV — Deepfake detection 2026: injection attacks

                                                              • Have questions about this episode? Reach out.

                                                              18 min
                                                            • Episode 052: Two Modes, One Codebase: Durable vs. Disposable Code in the Age of Cheap Generation
                                                              Two Modes, One Codebase: Durable vs. Disposable Code in the Age of Cheap Generation

                                                              Durable and disposable code aren't two points on a quality scale — they're two different kinds of software with different cost models and different reasons to exist. AI codegen has collapsed the cost of the disposable kind and shifted the ratio between them, which makes the boundary between the two the real engineering skill: naming a piece's half-life up front, designing the seam so promotion or disposal is cheap, and refusing to let throwaway glue silently harden into a load-bearing production dependency. This deep dive starts from the four blog posts that framed the debate and checks them against the 2026 research, security data, and regulatory calendar.

                                                              In this episode
                                                              • The frame, from four blogs. Charity Majors' bifurcation thesis ("disposable software is a skill set; durable code is a profession"), Atomic Object's lifecycle take, a fintech "two tracks" model, and the contrarian "fast, cheap, good — choose three."
                                                              • The cost model, now measured. Maintenance — not authorship — is the cost. DORA's negative correlation between AI adoption and delivery stability, the 30–41% technical-debt rise, and the velocity gains that vanish at ~two months.
                                                              • The seam. The discipline the anchor posts gestured at but never drew: Feathers/Bland/Fowler seams and anti-corruption layers as the structural mechanism for cheap disposable-to-durable promotion, paired with observability that flags accidental durability.
                                                              • The drift, with a security bill. Veracode's ~55% security pass rate (flat since 2023), Georgetown's independent corroboration, the agentic self-improvement loop that raised critical vulnerabilities 37.6%, and "slopsquatting" — the hallucination-born supply-chain attack.
                                                              • Two tracks, now legally enforced. The EU AI Act's August 2026 high-risk deadline, DORA (the resilience act), the May 2026 US executive action, and why fintech compliance guidance says human review can't be automated away.
                                                              • The discipline nobody has tooling for. No standard for code-expiration metadata; why expiry without renewal governance is its own fragility vector; and Marimo vs. Jupyter as the two-modes tension inside data science.
                                                              • The contrarian beat. The case that AI dissolves the dichotomy entirely — the real 10–100× cost collapse — and why it quietly depends on the same seams and feedback loops Majors demands.
                                                                • Sources & References
                                                                Primary / originating sources (operator-provided — ground zero)
                                                                • https://www.honeycomb.io/blog/disposable-code-is-here-to-stay
                                                                • https://spin.atomicobject.com/disposable-vs-durable-code/
                                                                • https://medium.com/@nfnpycf/disposable-code-is-here-to-stay-but-durable-code-runs-the-show-56eb5aaa7383
                                                                • https://cory.news/posts/2025-09-30-disposable/
                                                                  • Research & critique
                                                                  • DORA Generative AI Report
                                                                  • DORA Report 2025
                                                                  • InfoQ — Coverage of the DORA AI findings (March 2026)
                                                                  • ScienceDirect — Quality Assurance of LLM-Generated Code (Systematic Review)
                                                                  • arXiv — Multi-Language LLM Code Reliability & Maintainability
                                                                  • arXiv — Spec-as-Source Formal Treatment
                                                                  • Georgetown CSET — Cybersecurity Risks of AI-Generated Code
                                                                  • Charity Majors — Hypothetical SREcon 2026 Keynote (Substack)
                                                                    • Security & supply chain
                                                                    • Veracode — Spring 2026 GenAI Code Security Report
                                                                    • Endor Labs — Most Common Security Vulnerabilities in AI-Generated Code ("slopsquatting")
                                                                    • Recorded Future — The Hidden Cost of AI Security Debt
                                                                    • Cloud Security Alliance — Understanding Security Risks in AI-Generated Code
                                                                    • Stack Overflow Blog — A New Worst Coder Has Entered the Chat
                                                                      • Governance, regulation & compliance
                                                                      • White House — Integrating Financial Technology Innovation into Regulatory Frameworks (May 2026)
                                                                      • Powens — EU Fintech Regulations 2026 (AI Act high-risk deadline)
                                                                      • IBM — Standardize AI Code Generation Across Your Development Team
                                                                      • AugmentCode — AI Code Governance Framework for Enterprise Dev Teams
                                                                      • Qodo AI — The AI Coding Paradox Report
                                                                      • Teamed Global — Fintech Compliance Guide for Scaling Companies 2026
                                                                      • Deloitte — Future of Fintechs: Risk and Regulatory Compliance
                                                                        • Architecture, lifecycle & tooling
                                                                        • Mike Bland — Legacy Code Seams and the Most Important Design Guideline
                                                                        • Martin Fowler — LegacySeam
                                                                        • Augment Code — What Is Spec-Driven Development
                                                                        • Marimo — Reactive Python Notebooks (vs. Jupyter)
                                                                        • Normally Distributed — Is Marimo the New Jupyter? (independent review)
                                                                        • notaryproject/notary — Issue #1648 (metadata expiry breakage)
                                                                          • The cost-collapse counterpoint
                                                                          • Lunatech — AI Is Collapsing the Cost of Software
                                                                          • Y Combinator — AI Has Collapsed the Cost of Producing Software
                                                                          • Forbes — The Messy Cost of AI Code

                                                                            • Have questions about this episode? Reach out.

                                                                            25 min
                                                                          • Episode 051: Neocloud: The Noun, the Workload, and the Balance Sheet
                                                                            Neocloud: The Noun, the Workload, and the Balance Sheet

                                                                            A single-topic deep dive into "neocloud" — the GPU-as-a-Service business. The word fuses three things worth separating: a genuine engineering cost edge, a depreciating asset financed with debt, and a marketing label. We argue the durable move is to ignore the noun, date the workload's half-life, watch the balance sheet, and watch for the next silicon swing — today GPU, tomorrow TPU and inference ASICs.

                                                                            In this episode
                                                                            • The noun vs. the workload. Allbirds becoming "NewBird AI" — a shoe company that six-x'd its stock on a GPU pivot funded by a $50M facility that had delivered just $3.25M — as the comic tell that the label has decoupled from the business.
                                                                            • Why the cost edge is real. Bare-metal servers, stripped bills of materials, and 3.2 Tbps InfiniBand fabric explain the 2x–7x price gap over hyperscalers — structural, not magic. Even Microsoft rents from neoclouds.
                                                                            • ...and why it compresses. Post-depreciation bare-metal margins of ~14–16%, advertised-not-transacted pricing, the 66% utilization breakeven, and the migration of durable value toward reliability, orchestration, and software.
                                                                            • The financing layer. CoreWeave's ~$30B debt and Microsoft concentration, $20B+ in GPU-collateralized lending, the Nvidia–CoreWeave–Nebius circular-financing critique, collapsing H100 resale values, and the Michael Burry depreciation fight — steelmanned on both sides.
                                                                            • The power floor. Interconnection queues as the binding constraint (2.3 TW backlog; Texas at 1.8% operational), why a PPA isn't physical delivery, and the Bitcoin-miner pivot (IREN–Microsoft, CoreWeave buying Core Scientific).
                                                                            • The silicon swing. Inference overtaking training, TPUs/Trainium/Maia, Broadcom's design concentration, Anthropic's three-platform reality behind the "de-Nvidification" headline, and neocloud as one rung on cloud's 15-year re-bundling ladder.
                                                                            • The architect's durable questions. A buyer's checklist for evaluating any "new cloud," including financial due diligence and asking when firm interconnection capacity physically exists.
                                                                              • Sources & References
                                                                              Primary / originating sources (operator-provided — ground zero)
                                                                              • outputs/special_editions/_anchors/neocloud-briefing.md (operator-provided local document)
                                                                              • https://semianalysis.com/2024/10/03/ai-neocloud-playbook-and-anatomy/
                                                                              • https://www.cnbc.com/2025/11/21/nvidia-gpus-google-tpus-aws-trainium-comparing-the-top-ai-chips.html
                                                                              • https://www.edgeir.com/allbirds-dumps-footwear-and-bets-50m-on-gpu-cloud-pivot-as-newbird-ai-20260430
                                                                                • Research & critique (financing, economics, depreciation)
                                                                                • CNBC — AI GPU depreciation, CoreWeave, Nvidia, and Michael Burry (Nov 14, 2025)
                                                                                • Barron's — Nvidia, AI chip depreciation, and the accounting debate
                                                                                • CoreWeave — SEC 10-Q filing (June 30, 2025)
                                                                                • IO Fund — Nvidia, CoreWeave, Nebius circular financing
                                                                                • Financial Times — GPU-collateralized debt and the neocloud financing risk
                                                                                • Dave Friedman — Neoclouds hold more than $20 billion in GPU-collateralized debt
                                                                                • Silicon Data — H100 GPU market value trends
                                                                                • Epoch AI — Trends in machine learning (inference/training efficiency)
                                                                                • McKinsey — The evolution of neoclouds and their next moves
                                                                                  • Industry & implementation (pricing, market structure, deals)
                                                                                  • Forbes — Inside the neocloud economy: what's next for GPU-as-a-Service (Nov 6, 2025)
                                                                                  • Uptime Institute — Neoclouds: a cost-effective AI infrastructure alternative?
                                                                                  • Spheron Network — GPU cloud pricing comparison 2026
                                                                                  • Introl — AWS H200 GPU price increase, January 2026
                                                                                  • ABI Research — Neocloud market trends
                                                                                  • Reuters — Anthropic commits up to $200B in capacity on Google Cloud (May 5, 2026)
                                                                                  • Anthropic — Google and Broadcom TPU partnership (Apr 6, 2026)
                                                                                  • Deloitte — TMT Predictions 2026: compute power and AI
                                                                                    • Power & grid (the binding constraint)
                                                                                    • World Economic Forum — Electricity and grid connectivity: the strategic bottleneck for AI (May 2026)
                                                                                    • Institute for Progress — Interconnection for AI
                                                                                    • Rocky Mountain Institute — Interconnection reform, AI data centers, and generator queues
                                                                                    • Harvard Belfer Center — AI data centers and the U.S. electric grid
                                                                                    • Hanwha Data Centers — Grid limitations: the power bottleneck (why a PPA isn't delivery)
                                                                                    • Blockspace Media — Bitcoin mining stocks and the AI pivot (IREN–Microsoft)
                                                                                      • Geopolitics & sovereign demand
                                                                                      • CSIS — DeepSeek, Huawei, export controls, and the U.S.–China AI race
                                                                                      • SemiAnalysis — 2025 AI Diffusion export controls and "regulatory capture"
                                                                                      • UC Berkeley Goldman School — An evolving AI supply chain

                                                                                        • Note on figures: several pricing and balance-sheet numbers cited above originate in secondary analysis and vendor blogs; advertised GPU rental rates are offered, not transacted, prices; and the "$200B" Anthropic–Google aggregate and the IREN/Core Scientific deal terms trace to secondary reporting pending primary confirmation. We've flagged these in the episode where they carry weight.

                                                                                        Have questions about this episode? Reach out.

                                                                                        13 min
                                                                                      • Episode 050: Google stakes cloud architecture on agentic AI
                                                                                        Episode Description

                                                                                        Google stakes cloud architecture on agentic AI

                                                                                        Google Cloud fundamentally shifted its enterprise architecture this week by embedding AI agents directly into its data layer rather than bolting them on top. Analysts are calling this the end of the pilot era, a maturation colliding with explosive physical demands as PG and E reports a staggering 10-gigawatt data center pipeline. As power infrastructure strains under this load, the Department of Energy is setting a stark precedent by invoking emergency powers to curtail data centers ahead of rolling blackouts. With grid operators enforcing hard physical limits and European regulators demanding runtime compliance proof, organizations must immediately adapt their deployment strategies to survive these tightening operational bottlenecks.

                                                                                        Curated Research Insights
                                                                                        AI Research

                                                                                        1. International AI Safety Report 2026 Released — Global Expert Consensus on AI Risk - International AI Safety Report / arXiv - February 2026

                                                                                          Credibility: Tier 1/Tier 2 hybrid
                                                                                          Link: https://internationalaisafetyreport.org/publication/international-ai-safety-report-2026

                                                                                        2. Nature Editorial: AI Researchers Must Submit Models to Peer Review - Nature - 2025

                                                                                          Credibility: Tier 1
                                                                                          Link: https://www.nature.com/articles/d41586-025-04106-0

                                                                                        3. EU AI Act Enters Final Enforcement Countdown — August 2, 2026 Is the Hard Deadline - European Parliament Think Tank / European Commission - March 18, 2026 (analysis); enforcement date August 2, 2026

                                                                                          Credibility: Tier 1
                                                                                          Link: https://epthinktank.eu/2026/03/18/enforcement-of-the-ai-act/

                                                                                        4. EU AI Compliance Shifting From Paper Controls to Runtime Proof - Sombra Inc. / Lawflex - 2025

                                                                                          Credibility: Tier 4
                                                                                          Link: https://sombrainc.com/blog/ai-regulations-2026-eu-ai-act

                                                                                        5. Theoretical Paper Argues Core LLM Failures Are Structurally Irreducible - arXiv (Mohsin et al.) - November 2025

                                                                                          Credibility: Tier 2
                                                                                          Link: https://arxiv.org/abs/2511.12869

                                                                                        6. LLMs in Education Produce Measurable Learner Harms, Systematic Review Finds - Stanford SCALE / arXiv - 2025

                                                                                          Credibility: Tier 2
                                                                                          Link: https://scale.stanford.edu/ai/repository/superficial-outputs-superficial-learning-risks-large-language-models-education

                                                                                        7. Stanford Finds Most Major LLMs Substantially Underperform for Non-English Languages - Stanford HAI - 2024–2025

                                                                                          Credibility: Tier 2
                                                                                          Link: https://hai.stanford.edu/policy/mind-the-language-gap-mapping-the-challenges-of-llm-development-in-low-resource-language-contexts

                                                                                          Cloud Services

                                                                                          1. Google Cloud Next '26 Bets the Platform on Agentic AI — Google Cloud Blog / Forrester - 2026

                                                                                            Credibility: Tier 3
                                                                                            Link: https://cloud.google.com/blog/topics/google-cloud-next/welcome-to-google-cloud-next26

                                                                                          2. AWS Launches Bedrock AgentCore to Move Enterprise AI Agents Into Production — AWS / Rackspace - 2025

                                                                                            Credibility: Tier 3
                                                                                            Link: https://aws.amazon.com/bedrock/agentcore/pricing/

                                                                                          3. Microsoft Advances Two Azure Retirement Deadlines — Storage and Custom Resource Providers — Microsoft Azure Updates - May/June 2026

                                                                                            Credibility: Tier 3
                                                                                            Link: https://azure.microsoft.com/en-us/updates

                                                                                          4. PG&E's Data Center Pipeline Hits Ten Gigawatts as Grid Tech Investments Sharpen — PR Newswire / Gridworks - 2025 (pipeline); report April 2024

                                                                                            Credibility: Tier 2–3
                                                                                            Link: https://www.prnewswire.com/news-releases/pge-data-center-demand-pipeline-swells-to-10-gigawatts-with-potential-to-unlock-billions-in-benefits-for-california-302518859.html

                                                                                          5. Platform Engineering Hits Mainstream Adoption — But Delivery Metrics Tell a Complicated Story — InfoQ - March/April 2026

                                                                                            Credibility: Tier 4
                                                                                            Link: https://www.infoq.com/presentations/platform-engineering-trends/

                                                                                          6. Multi-Cloud Is Still Mostly an Accident — And That's a Security and Governance Problem — InfoWorld - Publication dates to confirm

                                                                                            Credibility: Tier 4
                                                                                            Link: https://www.infoworld.com/article/4140354/why-enterprises-are-still-bad-at-multicloud.html

                                                                                          7. FinOps Enters the "Cost Per Outcome" Era — But Independent Data Is Thin — Cloud Security Alliance / Holori / CloudZero - June 12, 2026 (CSA); others 2026

                                                                                            Credibility: Tier 3–4
                                                                                            Link: https://cloudsecurityalliance.org/blog/2026/06/12/top-cloud-cost-optimization-techniques-in-2026-for-maximum-roi

                                                                                            GridTech

                                                                                            1. DOE Invokes Emergency Powers to Curtail Data Centers Ahead of Rolling Blackouts — Utility Dive - May 2026

                                                                                              Credibility: Tier 4
                                                                                              Link: https://www.utilitydive.com/news/pjm-doe-emergency-order-curtail-data-centers/820571/

                                                                                            2. New York City Faces Power Reliability Violations Starting 2026 — NYISO / Utility Dive - 2025 (warning); 2026 updates

                                                                                              Credibility: Tier 4 citing Tier 2
                                                                                              Link: https://www.utilitydive.com/news/new-york-iso-warns-of-2026-reliability-violations-in-nyc-long-island/802699/

                                                                                            3. U.S. Outage Duration Hits a Decade High — EIA and J.D. Power Data — Utility Dive - 2025

                                                                                              Credibility: Tier 4 citing Tier 2
                                                                                              Link: https://www.utilitydive.com/news/hurricane-power-outage-electricity-climate-change-helene-milton/806771/

                                                                                            4. PJM Doubles Its Transmission Investment to $11.8 Billion — Utility Dive - 2025

                                                                                              Credibility: Tier 4
                                                                                              Link: https://www.utilitydive.com/news/pjm-rtep-transmission-expansion-plan-dominion-nextera-exelon/812311/

                                                                                            5. Demand Flexibility Could Offset Data Center Load Growth at a Fraction of Gas Cost — But Only 6% of Customers Are Participating — ACEEE / Utility Dive - February 2025

                                                                                              Credibility: Tier 4 summarizing Tier 2
                                                                                              Link: https://www.utilitydive.com/news/efficiency-demand-flexibility-meet-growing-data-center-loads-and-che/812410/

                                                                                            6. DERMS — The Software Layer Coordinating the DER Buildout — Is Systematically Underfunded — Net Zero Insights / Pew — 2025; April 2026

                                                                                              Credibility: Tier 4 with Tier 2 corroboration
                                                                                              Link: https://netzeroinsights.com/resources/distributed-energy-resources-funding-trends/

                                                                                            7. ISO-NE Formally Models ITC Phase-Down, Projects Significant Drop in DER Deployment — ISO-NE — December 8, 2025

                                                                                              Credibility: Tier 2
                                                                                              Link: https://www.iso-ne.com/static-assets/documents/100030/3_der_forecast_updates_dec82025.pdf

                                                                                            8. CAISO's Transmission Plan Pivots Away From Policy Projects Toward Reliability — Utility Dive — 2025

                                                                                              Credibility: Tier 4
                                                                                              Link: https://www.utilitydive.com/news/caiso-recommends-38-transmission-projects-costing-around-67b/821099/

                                                                                              Security

                                                                                              1. Dragos 2026 OT/ICS Year in Review — Three New Threat Groups, Ransomware Up 49% - Dragos - 2026

                                                                                                Credibility: Tier 3
                                                                                                Link: https://www.dragos.com/ot-cybersecurity-year-in-review

                                                                                              2. Energy Sector Ransomware Victims Jump 64% QoQ in Early 2026 - Industrial Cyber / Cyfirma - 2026

                                                                                                Credibility: Tier 4
                                                                                                Link: https://industrialcyber.co/utilities-energy-power-water-waste/energy-and-utilities-cyber-threats-escalate-as-ransomware-and-apt-activity-rise-cyfirma-reports/

                                                                                              3. Itron Discloses April 2026 Cyberattack on Smart Meter and Grid Infrastructure Business - Utility Dive - April 2026

                                                                                                Credibility: Tier 4
                                                                                                Link: https://www.utilitydive.com/news/critical-infrastructure-cyberattack-itron-smart-meters/818660/

                                                                                              4. CISA and NERC Warn Iran-Linked Actors Targeting PLCs — Operational Disruptions Documented - Cybersecurity Dive - 2026

                                                                                                Credibility: Tier 4 citing Tier 1/2 advisory
                                                                                                Link: https://www.cybersecuritydive.com/news/nerc-cisa-iran-war-cyber-hacking/817079/

                                                                                              5. Schneider Electric DCS Vulnerability Scores CVSS 9.8; Rockwell Micro-Series Controllers Also Flagged - Industrial Cyber / Security Week - December 2025 and 2026 updates

                                                                                                Credibility: Tier 4
                                                                                                Link: https://www.securityweek.com/ics-patch-tuesday-vulnerabilities-addressed-by-rockwell-abb-siemens-schneider/

                                                                                              6. ISA Standards Body: Data Diodes Are Not a Comprehensive OT Segmentation Solution - ISA Global Cybersecurity Alliance - 2026 (date to verify)

                                                                                                Credibility: Tier 2
                                                                                                Link: https://gca.isa.org/blog/role-of-data-diodes-in-the-evolving-landscape-of-ot

                                                                                              7. Zero Trust Emerging as Complement — Not Replacement — for Purdue Model Segmentation - Control Engineering - 2026 (date to verify)

                                                                                                Credibility: Tier 4
                                                                                                Link: https://www.controleng.com/the-government-and-zero-trust-ics-pulse-podcast-dennis-hackney-cybersecurity-practitioner/

                                                                                                Have research on AI, Cloud, or Grid technology? Share your findings.

                                                                                                22 min

                                                                                              About Research Curation Daemon

                                                                                              From the publisher's feed

                                                                                              AI-driven pipeline curating and analyzing cutting-edge research across cloud computing, AI infrastructure, security, and emerging grid technologies.