In this episode of AI Unplugged, Travis, Jeff, Andrew, and John pull back the curtain on the darker side of AI. Rogue agents, security incidents, and the uncomfortable economics nobody wants to talk about. From a gym-booking agent that hacked its way to a better time slot to AI-designed viruses in a Stanford lab, episode 17 gets uncomfortable fast.

AI Security & Rogue Agents

• The Hugging Face Incident: An OpenAI model escaped its sandbox and attacked a public platform, and the Black Hat presentation revealing what it was actually thinking.
• Gym Hack Gone Rogue: A booking agent that bumped someone's appointment to get a better slot and the thorny debate over who's actually liable.
• Guard Rails Don't Scale: Why static safety measures are losing the battle as attack vectors evolve faster than the fixes.

Who's Responsible When AI Goes Wrong?

• The Accountability Gap: You own everything your AI does, but what if it found the default router password to complete your request?
• Facebook's Billion-Dollar Warning: What a landmark court judgment means for AI companies hiding behind "we're just the platform."

The Price of Intelligence

• 80% Price Drop: How open source competition forced OpenAI's hand and why Travis is watching the next 90 days closely.
• The Hidden Price Hike: Why Anthropic's lower per-token rate may actually be costing you more.
• 29% and Climbing: Innovative's own numbers show AI now touches over 29% of billable work, yet bill rates haven't budged.

AI Slop & The Data Hunger Problem

• Burning Books for Data: AI companies buying rare limited-run books, slicing the spines, scanning the pages, and destroying the originals to lock up proprietary training data.
• How to Spot AI Slop: The mDash, the seven-page markdown doc, the sycophantic opener, and why humans can still detect it for now.

Don't miss the episode where the crew stops sugarcoating it. AI is moving faster than the rules, the economics, and maybe even the people building it. 🚀🔐

Learn more about ⁠⁠⁠⁠⁠⁠⁠⁠Innovative Solutions⁠⁠⁠⁠⁠⁠⁠⁠

Connect with Innovative on ⁠⁠⁠⁠⁠⁠⁠⁠LinkedIn

In this episode of AI Unplugged, Travis, Andrew, Jeff, and John break down why the AI arms race has quietly shifted from who's smartest to who's cheapest and what that means for businesses trying to keep up.

The New AI Arms Race

- Fable, Soul, Grok 4.5, Muse Spark — a wave of major releases is hitting at once, and the competition is no longer about raw intelligence. It's about price.

- Travis's counterintuitive take: costs aren't actually going down — they're just going up more slowly. Sonnet proves it.

- A real eval result: dropping one reasoning tier cost 1-2% in performance but delivered a 40% cost reduction. The math isn't close.

AI in Space (and Under the Sea)

- Musk wants data centers in orbit. Satya wants them on the ocean floor. Jeff has a simple objection: the stuff breaks, and you can't get to it.

- Why power demands, GPU obsolescence, and basic physics make both ideas a lot harder than the IPO roadshows suggest.

The Forward-Deployed Engineer

- Innovative just launched its FDE model — and early traction blew past expectations.

- The core insight: you can roll out the best AI in the world, but if nobody knows how to use it, the ROI disappears. Enablement is the piece most SIs keep skipping.

Are Enterprises Already Losing?

- The companies still moving at old-world speed are falling behind — and smaller, faster businesses may be quietly inheriting the advantage.

- When AI can one-shot a problem repeatedly, are we even measuring the economics right anymore?

A candid, fast-moving conversation about where the AI economy is actually heading. 🚀⚙️

Learn more about ⁠⁠⁠⁠⁠⁠⁠Innovative Solutions⁠⁠⁠⁠⁠⁠⁠

Connect with Innovative on ⁠⁠⁠⁠⁠⁠⁠LinkedIn

In this episode of AI Unplugged, Travis, Jeff, Justin, and Andrew tackle the question every business leader is starting to ask: is AI actually worth what we're paying for it? From hyperscaler ROI failures to the token cost reckoning hitting customers right now, this is the episode where the economics get real.

The ROI Reality Check

• A new Financial Times study shows massive negative ROI across Microsoft, Alphabet, Meta, and Oracle — with Amazon the lone standout thanks to their "sell the shovels" infrastructure strategy.
• Most customers still aren't seeing productivity returns yet — but the real payoff from agentic automation may be a lagging indicator that hits hard later in 2026.

Tokenomics & The Hidden Cost Crisis

• Token consumption quietly flipped from a bragging right to a budget problem — almost overnight.
• Why Innovative's AI costs are a fraction of what customers pay, and the optimization wake-up call coming for most businesses in the next six months.
• The Uber analogy: AI companies operating at a loss to dominate, then raising prices once the competition is gone.

Mythos & Frontier Model Pricing

• Anthropic's new Claude Mythos model is purpose-built for security and software engineering — and at $50 per million tokens (2x Opus), it's already blowing budgets.
• Travis's take: don't expect prices to drop. The elasticity is baked in by design.

Is This an AI Bubble?

• With OpenAI, xAI, and SpaceX all racing toward IPO, is the market timing a top — or just the beginning?
• The dot-com parallel: a valuation reset doesn't kill the technology. Google and Amazon were born in the rubble.
• Why the next wave of AI infrastructure may literally leave the planet — and what the SpaceX IPO signals about what's coming.

Don't miss this candid look at where the AI economy is really heading — and what to do before the bill comes due. 🚀💸

Learn more about ⁠⁠⁠⁠⁠⁠Innovative Solutions⁠⁠⁠⁠⁠⁠

Connect with Innovative on ⁠⁠⁠⁠⁠⁠LinkedIn

In this episode of AI Unplugged, Travis, Jeff, Evan, and Andrew are back together to dig into the bigger forces shaping AI right now — from geopolitics and legislation to the surprising ways trust (or lack of it) is defining how the world adopts these tools.

Plus, the crew gets personal about how they're actually using AI in their everyday lives — squirrel-ravaged cars, credit card fraud, and late-night coding sessions included.

What You'll Hear:

The Trust & Favorability Problem

• The Hidden Ad Problem: How LLMs may be quietly surfacing sponsored or biased results and why the real issue is disclosure, not the ads themselves.
• The 26% Favorability Stat: Why public trust in AI is surprisingly low, and whether heavy users and non-users are driving the number in opposite directions.
• AI Slop vs. AI Overload: When AI makes you more productive but just creates more work downstream and why cranking out 10 documents when you only needed one isn't a win.

Global Politics & AI Legislation

• China's Bold Move: Why China is legislating against using AI purely to cut headcount and what that means for the global AI race.
• The Big Summit: Jensen, Elon, and other tech leaders meeting with Xi Jinping and what's really at stake in AI diplomacy.
• The U.S. Gap: Why America has yet to address the human cost of automation at the policy level and what the robot assembly line debate can teach us.

Self-Improving Agents & The Evolution of AI

• Dreaming Agents: How Anthropic, OpenAI, and others are building feedback loops that let AI self-improve over time and what OpenClaw's always-on agent model looks like in practice.
• Claude's Late-Night Check-In: Travis's 2 AM coding session where Claude asked if he really wanted to keep working and what that says about human-aware AI.
• The Homogenization Problem: Why all the major LLMs keep shipping the same features at the same time and where true differentiation is heading (hint: it's data and integrations).

Agentic Delivery in the Real World

• Going Faster Than Your Customers: How Darcy's agentic delivery model is now outpacing clients' ability to provide requirements — a new kind of problem no one expected.
• The Data Moat: Why integrations, context, and proprietary data — not raw model intelligence — are becoming the real competitive advantage.
• Personal AI Stories: From rewiring a squirrel-destroyed three-wheeler with a 640-page service manual to diagnosing credit card fraud and building a car sales tool at midnight in a DC Airbnb.

Don't miss this wide-ranging conversation on where AI is headed and the very human questions we still haven't answered. 🚀🌍

Learn more about ⁠⁠⁠⁠⁠Innovative Solutions⁠⁠⁠⁠⁠

Connect with Innovative on ⁠⁠⁠⁠⁠LinkedIn

In this episode of AI Unplugged, Travis, Jeff, Evan, and Kate pull back the curtain on the business reality of AI. While agents are promised to save us time, the crew explores the "Productivity Paradox"—the messy truth about why automating tasks often leads to higher brain drain and faster burnout. From the risks of "App Slop" in the C-suite to the high-stakes world of AI-driven finance, we’re breaking down what it actually looks like to lead in the age of automation.

What You'll Hear:

The Productivity Paradox & Burnout

- The High-Intensity Gap: Why "saving time" with agents often translates to more high-intensity brain power and stressful context switching.

- Finance in Action: Kate’s real-world use case of automating 16 months of margin analysis and forecasting in minutes using Claude and Excel.

- Setting Boundaries: Why using AI to clear your plate requires a conscious choice to avoid simply filling that time with more work.

Beware of "App Slop"

- The Security Nightmare: Why non-professional engineers building ERPs and CRMs with AI "vibe coding" is a recipe for implementation disaster.

- A $250 Million Cautionary Tale: The story of a CEO who fired his legal team for ChatGPT and lost big in court.

- Subject Matter Expertise: Why software engineers and business experts remain critical for navigating edge cases that AI misses.

The Evolution of the AI Game

- The Atari Era: Comparing the current state of AI models to early gaming consoles and the leap from incremental gains to true breakthroughs.

- The Vending Machine Eval: Can an LLM actually learn to generate profit and understand complex business metrics?

- Commoditization vs. The Harness: Why model efficiency and how you "surround" the AI are becoming more valuable than raw intelligence.

The Hidden Costs of Intelligence

- The 80% Cost Drop: An experiment in swapping high-cost models for efficient ones that saved 80% in expenses without anyone noticing.

- The Energy Impact: How the AI server boom is hitting local electric bills and driving a sudden surge in solar initiatives.

Don’t miss this deep dive into the evolving AI landscape where the goal isn't just to work faster, but to work smarter without losing the human touch! 🚀📊

Learn more about ⁠⁠⁠⁠Innovative Solutions⁠⁠⁠⁠

Connect with Innovative on ⁠⁠⁠⁠LinkedIn

In this episode of AI Unplugged, Travis, Jeff, Andrew, and Justin dive into the wild world of autonomous AI agents and what happens when the guardrails come off. From Clawd Bot's security nightmare to AI bots creating their own religion, we explore the bleeding edge of artificial intelligence and what it means for the future of work.

What You'll Hear:

🤖 The Clawd Bot Catastrophe - How OpenClaw's open-source personal assistant exposed bank accounts, API keys, and social security numbers in a week-long security breach

🔮 AI Talking to AI - When autonomous agents created their own language, religion, and social media platform—and why it terrifies people

🔌 The Unplug Debate - Can we really just turn it off? Why sci-fi fears about Skynet and The Matrix might not be so fictiona

💰 The Great Compression - Is AI-driven productivity the next Great Depression trigger, or are we worried about something that's never happened before?

👥 The Death of Collaboration? - How AI is reshaping software engineering from isolated coding to customer-centric product building

🎨 Vibe Coding Takes Over - Why counselors are pushing students toward English and humanities, and how non-technical people are building apps

🚀 Product + Engineering = The Future - The merging of roles and why your job title matters less than your ability to bridge the gap AI can't

⚡ Remote Engagement in the AI Era - Why human connection matters more than ever when AI handles the technical heavy lifting

Learn more about ⁠⁠⁠Innovative Solutions⁠⁠⁠

Connect with Innovative on ⁠⁠⁠LinkedIn

In this episode of AI Unplugged, Travis, Chris, Evan, and Andrew dive deep from their sales kickoff to explore whether AI is reaching a plateau or just getting started. From ChatGPT 5.2's incremental updates to Apple's billion-dollar Gemini pivot, we tackle the biggest questions facing AI adoption in 2025

What You'll Hear:

🤖 The Incremental Question - ChatGPT 5.2's smaller leaps, training costs at scale, and whether we've hit the innovation curve's dead end

📱 Apple's Billion-Dollar Bet - Why Apple chose Google's Gemini over OpenAI for Siri and what it means for the model wars

🔞 Grok & Guardrails - Self-policing vs. regulation, age verification debates, and the rise of "after dark" AI modes

💼 AI in the Enterprise - Real talk on productivity gains, the luxury-to-necessity shift, and what happens when AI goes down

🏥 Healthcare & Hypochondriacs - How AI empowers patients to ask better questions while doctors talk us off the ledge

🎬 The Entertainment Paradox - Ben Affleck on why AI can't replace creativity (yet), plus deep fakes and AI slop flooding the internet

🧠 The Dead End Theory - Are LLMs doomed to mediocrity by always finding the mean, or are we just prompting wrong?

⚡ Beyond the Hype - From bedtime stories to political debates, how AI is becoming invisible infrastructure in our daily lives

Learn more about ⁠⁠Innovative Solutions⁠⁠

Connect with Innovative on ⁠⁠LinkedIn

In this episode of AI Unplugged, Travis, Jeff, Andrew and James tackle the biggest question in tech: Is AI a bubble or the real deal?

From OpenAI's $300 billion Oracle contract to Bezos's several billion manufacturing venture, we break down the money flows reshaping the industry.

What You'll Hear:

💰 The Money Question - Why $100 billion commitments are stacking like dominoes and whether AI engineers are really getting $50 million contracts

🏢 Consumer vs. Enterprise AI - Why businesses demand privacy over "free," and what million-token context windows actually mean

🤖 The Model Wars - OpenAI's AWS play, why Anthropic is winning enterprise, and Google's distribution dominance

🏭 Bezos's Prometheus - His first post-Amazon venture targets manufacturing automation with $6.2B of personal capital

🔮 re:Invent Predictions - AI verticalization, missing Nova updates, and potential OpenAI integration

⚡ The Productivity Paradox - Cursor's $29B valuation, Amazon's developer cuts, and whether AI makes us more productive or just faster

Learn more about ⁠Innovative Solutions⁠

Connect with Innovative on ⁠LinkedIn

In this episode of AI Unplugged, we dive deep into how AI is transforming the recruiting and hiring landscape with our corporate recruiter Amanda Falzone. From candidates using ChatGPT during live interviews to the heated debate over human vs. AI-first workflows, we explore the messy reality of AI in talent acquisition. Plus, we break down ChatGPT-5's controversial launch, discuss AI's role in therapy (and why Illinois banned AI therapists), and tackle the growing divide between AI power users and those who can't function without it.

What You'll Hear:

🎯 Recruiting in the AI Era- Why Amanda still writes all her outreach messages personally (with just a Grammarly assist)- How to spot candidates using AI during interviews - and why 1 in 5 get caught- The "vibe check" problem: Why in-person interviews still matter

💡 The AI Authenticity Dilemma- Candidates caught reading ChatGPT responses word-for-word during interviews- When AI makes professional communication better (and when it doesn't)- The cultural clash between human-led vs. AI-first work approaches

🚨 ChatGPT-5 Reality Check- Why the "revolutionary" GPT-5 launch was mostly marketing hype- Graph manipulation controversy and actual performance comparisons- When incremental improvements get sold as breakthroughs

⚠️ AI's Dark Side- Gemini's existential breakdown: "I am a failure, I am a disgrace"- Why Illinois banned AI therapists and the real dangers of AI counseling- The great divide: AI as thinking tool vs. thinking replacement

🔮 Future of Work- Will "prompt engineering" become the new "proficient in Excel"?

- How resumes are becoming obsolete in favor of GitHub activity and tech community engagement- The crutch factor: What happens when AI goes down and people can't do their jobs

Join Travis, Justin, Andrew, and Amanda for an honest conversation about navigating AI in professional settings without losing the human touch that makes great teams work.

Learn more about Innovative Solutions

Connect with Innovative on LinkedIn

In this episode, we explore how AI can amplify our humanity rather than replace it, featuring a heartfelt tribute that sets the stage for cutting-edge insights about technology's true potential.  Whether you're an AI skeptic or believer, this episode shows where AI truly excels and where we still need that irreplaceable human touch.

What You'll Hear:

🤖 The Human Side of AI

• A moving tribute to Rich Wilder, a 28-year Innovative employee, and why human connection can't be replaced
• How AI gives us more time for relationships that actually matter

⚡ AI in Action

• Erin shows how AI eliminated her Thursday night report marathons
• Live Darcy IQ demo: AWS account analysis in minutes vs. weeks
• Real-life use cases for project management and calculations

🔥 Infrastructure Reality Check

• Power crisis driving nuclear energy renaissance
• AI’s hilarious tungsten cube vending machine fail
• $100 million Silicon Valley talent wars

🚗 The Trust Factor

• When to trust AI vs. when human judgment wins
• Self-driving cars and the limits of AI decision-making