Society Tech Brief By HackerNoon

Society Tech Brief By HackerNoon

Download on the App Store

Society Tech Brief By HackerNoon episodes

  • K-EXAONE 2.0 Brings 262K Context to Frontier AI

    This story was originally published on HackerNoon at: https://hackernoon.com/k-exaone-20-brings-262k-context-to-frontier-ai.


    Explore K-EXAONE 2.0, LG’s 750B-parameter multilingual AI model with 262K context, dual reasoning modes, agentic coding, and Apache 2.0 licensing.
    Check more stories related to society at: https://hackernoon.com/c/society.
    You can also check exclusive content about #legal, #machine-learning, #performance, #algorithms, #api, #artificial-intelligence, #k-exaone-2.0, #k-exaone-750b, and more.


    This story was written by: @aimodels44. Learn more about this writer by checking @aimodels44's about page,
    and for more stories, please visit hackernoon.com.


    Explore K-EXAONE 2.0, LG’s 750B-parameter multilingual AI model with 262K context, dual reasoning modes, agentic coding, and Apache 2.0 licensing.

    20 min
  • The Future of the Internet May Depend on Proving You Are Human, Without Revealing Who You Are

    This story was originally published on HackerNoon at: https://hackernoon.com/the-future-of-the-internet-may-depend-on-proving-you-are-human-without-revealing-who-you-are.


    Bots now generate 53% of web traffic and AI fraud is surging. Inside privacy-preserving proof of human: World, Self Protocol, Humanity Protocol and what's next.
    Check more stories related to society at: https://hackernoon.com/c/society.
    You can also check exclusive content about #world, #world-id, #proof-of-human, #deepfakes, #web3, #good-company, #privacy, #worldcoin, and more.


    This story was written by: @ishanpandey. Learn more about this writer by checking @ishanpandey's about page,
    and for more stories, please visit hackernoon.com.


    Automated traffic reached 53% of all web activity in 2025, per Imperva's 2026 Bad Bot Report. Humans are now a structural minority on their own internet, and the share is still falling.
    The FBI's Internet Crime Complaint Center logged $893 million in AI-referenced fraud losses in 2025, its first year tracking AI as a standalone category, while Deloitte projects generative AI fraud losses of $40 billion in the US by 2027.
    What is privacy-preserving proof of human? It is a class of verification systems that confirm a user is a real, unique human being without collecting or revealing personal information such as names, government ID numbers or contact details, typically using zero-knowledge proofs paired with a uniqueness anchor like an iris scan, passport chip or palm print.
    How is proof of human different from KYC? KYC establishes who you are and stores identifying data with the verifier. Proof of human establishes only that you are a real and unique person, and the verifier learns nothing else about you.
    Why does the internet need proof of human now? Automated traffic passed 53% of all web activity in 2025, AI-referenced fraud losses reported to the FBI reached roughly $893 million in the same year, and humans detect deepfake video less than a quarter of the time, so platforms can no longer rely on behavioral signals to separate people from machines.
    Which companies offer privacy-preserving proof of human? World issues World IDs anchored to iris verification with nearly 18 million verified humans. Self Protocol generates zero-knowledge proofs from biometric passports, national IDs and Aadhaar. Humanity Protocol issues palm-anchored verifiable credentials. Adjacent tools include Apple's Private Access Tokens, Cloudflare's Privacy Pass and Google Wallet's zero-knowledge age assurance.
    Is proof of human safe for user privacy? The leading designs keep biometric data on the user's device and share only cryptographic proofs, but regulators in several countries, including the Philippines and Thailand, have challenged biometric collection practices, so jurisdiction-level answers still differ.
    How does proof of human work with AI agents? Emerging tools bind a verified human to the agents acting on their behalf. World's AgentKit lets merchant sites confirm a real human authorized an AI shopping agent's purchase, and per-human rate limiting lets services cap agent traffic by the person behind it rather than by IP or device.
    Will proof of human replace passwords or logins? Not directly. It complements authentication by answering a different question: not "is this the account owner" but "is this account backed by exactly one real human." Expect it to appear alongside existing logins first, as badges, gated rewards and abuse-protected tiers.
    Deepfake supply exploded from an estimated 500,000 files in 2023 to 8 million by 2025, while humans detect fake video less than a quarter of the time. Detection is losing; verification is the fallback.
    The old tradeoff, verify identity and lose privacy or stay anonymous and drown in abuse, is dissolving. A third category, privacy-preserving proof of human, verifies uniqueness and humanness while disclosing nothing else.
    World has crossed nearly 18 million Orb-verified humans and shipped integrations with Tinder, Zoom, Docusign, Razer and Okta. Self Protocol anchors humanness to government documents via zero-knowledge proofs and has been adopted by Google Cloud and Aave. Humanity Protocol bets on palm biometrics and counts Mastercard as a partner.
    The next 18 months decide whether proof of human becomes internet infrastructure or stays a Web3 niche. The signals to watch: platform abuse disclosures, agent identity standards, regulatory rulings on biometric custody, and whether verification survives its own black market.

    13 min
  • Fewer Clicks, More Answers: MileMark Is Preparing Law Firms for the New Search Landscape

    This story was originally published on HackerNoon at: https://hackernoon.com/fewer-clicks-more-answers-milemark-is-preparing-law-firms-for-the-new-search-landscape.


    MileMark Legal Marketing, an agency that works exclusively with law firms, began preparing for this shift well before it became a standard talking point
    Check more stories related to society at: https://hackernoon.com/c/society.
    You can also check exclusive content about #legal, #milemark-legal-marketing, #aeo-for-lawyers, #rank-legal-service-in-chatgpt, #rank-legal-service-in-gemini, #rank-legal-service-in-google, #technologywire-press-release, #good-company, and more.


    This story was written by: @technology_wire. Learn more about this writer by checking @technology_wire's about page,
    and for more stories, please visit hackernoon.com.


    Language models favor sources that are clearly structured, factually consistent across the web, and demonstrably authoritative on a specific subject. In practice that means schema markup identifying attorneys, services, and service areas in machine readable form. It means content written to answer real questions directly rather than circling them for four paragraphs before arriving at the point. It means firm details that match across directories, bar listings, review platforms, and social profiles, since contradictions quietly undermine confidence in a source.

    4 min
  • #FreeYakutia: How a Viral Meme Turned Into a Conversation About Independence

    This story was originally published on HackerNoon at: https://hackernoon.com/freeyakutia-how-a-viral-meme-turned-into-a-conversation-about-independence.


    If Russia constantly talks about controlling other nations, what happens if people start discussing the independence of regions inside Russia itself?
    Check more stories related to society at: https://hackernoon.com/c/society.
    You can also check exclusive content about #ukraine, #russia-ukraine-war, #history, #independence, #china, #politics, #usa, #hackernoon-top-story, and more.


    This story was written by: @arthurtkachenko-rm2m8. Learn more about this writer by checking @arthurtkachenko-rm2m8's about page,
    and for more stories, please visit hackernoon.com.


    #FreeYakutia. It was partly satire, partly political trolling, and partly genuine curiosity about regions most people outside Russia knew almost nothing about.

    20 min
  • Facebook Marketplace Was Built for Used Couches. US Dropshippers Are Using It Anyway

    This story was originally published on HackerNoon at: https://hackernoon.com/facebook-marketplace-was-built-for-used-couches-us-dropshippers-are-using-it-anyway.


    Facebook Marketplace is evolving beyond local classifieds into a growing sales channel for US dropshippers seeking alternatives to Amazon's stricter ecosystem.
    Check more stories related to society at: https://hackernoon.com/c/society.
    You can also check exclusive content about #facebook-marketplace, #us-dropshipping-suppliers, #amazon-seller-alternatives, #meta-marketplace-sellers, #domestic-dropshipping-usa, #marketplace-ecommerce, #dropshipping-platform, #good-company, and more.


    This story was written by: @sanya_kapoor. Learn more about this writer by checking @sanya_kapoor's about page,
    and for more stories, please visit hackernoon.com.


    Originally built for local secondhand sales, Facebook Marketplace is attracting US dropshippers looking to diversify beyond Amazon. With nationwide shipping, direct buyer communication, and lower barriers to entry, the platform is becoming a viable ecommerce channel. Success increasingly depends on fast domestic fulfillment, making US supplier networks like Doba a competitive advantage.

    6 min
  • The Next Big Social Media Platform Won't Look Like the Current Ones

    This story was originally published on HackerNoon at: https://hackernoon.com/the-next-big-social-media-platform-wont-look-like-the-current-ones.


    A year of conversations with people reveals the same frustrations with algorithmic feeds, misinformation, and lost privacy.
    Check more stories related to society at: https://hackernoon.com/c/society.
    You can also check exclusive content about #social-network, #social-media, #decentralized-social-media, #algorithms, #privacy, #misinformation, #future-of-social-media, #tech-opinion, and more.


    This story was written by: @iambatman. Learn more about this writer by checking @iambatman's about page,
    and for more stories, please visit hackernoon.com.


    After noticing his own compulsive, unsatisfying relationship with social media, the author spent a year deliberately asking people online and in person, across ages and backgrounds whether they felt it too. Nearly everyone did, describing the same handful of problems: addictive design, unexplained irritation, an inability to tell real from fake (especially with AI), a desire for feeds they control, disappearing privacy as data trains AI models, and safety treated as an afterthought. The piece pairs each observation with current research and data, then looks honestly at where platforms newer emerging alternative platforms still fall short of actually solving it.

    13 min
  • Flex Raises $70M Led by Halo Fund to Scale AI-Native Private Banking for Global Business Owners

    This story was originally published on HackerNoon at: https://hackernoon.com/flex-raises-$70m-led-by-halo-fund-to-scale-ai-native-private-banking-for-global-business-owners.


    Flex raises $70M led by Halo Fund at a reported $1.2B to scale its AI-native private bank for middle-market business owners and launch Flex Global.
    Check more stories related to society at: https://hackernoon.com/c/society.
    You can also check exclusive content about #flex, #halo-fund, #stablecoins, #good-company, #web3, #artificial-intelligence, #portage, #startup-funding, and more.


    This story was written by: @ishanpandey. Learn more about this writer by checking @ishanpandey's about page,
    and for more stories, please visit hackernoon.com.


    Flex raised $70 million in a Series B1 led by Halo Fund, the $1 billion Utah vehicle co-founded by Qualtrics founder and Utah Jazz owner Ryan Smith with Accel's Ryan Sweeney. Portage, Wellington, Crosslink Capital, 53 Stations, Titanium Ventures, Spice and Florida Funders joined. Reuters sources the valuation at roughly $1.2 billion.
    Annualised payment volume crossed $10 billion, roughly 4x year on year. Revenue is up 3x since December 2025, on a nine-figure run rate. Total equity is now $180 million against $300 million of debt.
    Flex is the cheapest fast-growing asset in business finance. At a nine-figure run rate, $1.2 billion implies roughly 5 to 10 times revenue. Ramp trades at 29 to 44 times. Airwallex at 8.5. Mercury at 8. Flex is growing faster than all three and priced below all three.
    Its take rate is the highest in the peer set. Flex earns 1.0% to 2.5% per dollar moved. Airwallex earns 0.45%. Mercury earns 0.26%. That is what a multi-product owner relationship buys.
    Flex Global launches stablecoin settlement in 100+ countries, multi-currency accounts across 76 countries and 32 currencies, institutional dollar accounts and private credit in 20+ markets. It cuts the cost of a cross-border dollar for the owner by up to 97%.
    The real asset is the data. Flex sees the company ledger and the owner's personal ledger at the same time, in a segment with no public credit market. Nobody else has both sides of that balance sheet.

    30 min
  • The Morning I Finally Looked Up at the Bridge

    This story was originally published on HackerNoon at: https://hackernoon.com/the-morning-i-finally-looked-up-at-the-bridge.


    A real-world lesson in reinforced concrete, bridge inspection, and structural engineering from discovering concrete spalling on my daily commute
    Check more stories related to society at: https://hackernoon.com/c/society.
    You can also check exclusive content about #civil-engineering, #structural-engineering, #infrastructure, #bridge-engineering, #structural-health-monitoring, #structural-damage, #reinforced-concrete, #infrastructure-damage, and more.


    This story was written by: @kamalsamaila. Learn more about this writer by checking @kamalsamaila's about page,
    and for more stories, please visit hackernoon.com.


    I thought a damaged bridge was on the verge of failure after spotting rusted, exposed reinforcement. Learning how reinforced concrete actually behaves taught me that visible damage isn't the same as structural failure—and that good engineering starts with asking better questions.

    5 min
  • From Drone Footage to 3D Battlefield Twin in Real Time: How Farsight Vision Is Rebuilding ISR

    This story was originally published on HackerNoon at: https://hackernoon.com/from-drone-footage-to-3d-battlefield-twin-in-real-time-how-farsight-vision-is-rebuilding-isr.


    Farsight Vision building a tool for 3D terrain reconstruction, object detection and mission planning
    Check more stories related to society at: https://hackernoon.com/c/society.
    You can also check exclusive content about #ukraine, #technology, #3d, #3d-mapping, #image-processing, #image-recognition, #drone-feed-ingestion, #3d-terrain-reconstruction, and more.


    This story was written by: @arthurtkachenko-rm2m8. Learn more about this writer by checking @arthurtkachenko-rm2m8's about page,
    and for more stories, please visit hackernoon.com.


    From drone feed ingestion to 3D terrain reconstruction, object detection, mission planning, and GPS-denied navigation

    8 min
  • What Building a Self-Paced Math System Taught Me About Software Design

    This story was originally published on HackerNoon at: https://hackernoon.com/what-building-a-self-paced-math-system-taught-me-about-software-design.


    An adaptive math platform reveals why dependency graphs, observability, and verification matter as much in education as they do in software engineering.
    Check more stories related to society at: https://hackernoon.com/c/society.
    You can also check exclusive content about #edtech, #ai-in-education, #adaptive-learning-systems, #mastery-based-learning, #dependency-graph-learning, #ai-tutoring, #edtech-software-design, #concept-dependency-mapping, and more.


    This story was written by: @matthewyip. Learn more about this writer by checking @matthewyip's about page,
    and for more stories, please visit hackernoon.com.


    I built an automated math learning system called Mathewmatician's Dictionary, and the deeper I went, the more it stopped feeling like an education problem and started looking like a software design problem. Three lessons carried over directly. First, when a student fails a topic, the real cause is almost always an unmastered prerequisite several chapters back, the same way a UI bug often lives in the data layer. Concepts have dependencies, like a build graph, so I stopped organizing learning by chapter and started organizing it by concept. Second, mastery before movement is just test-driven development for humans: a concept has to pass its own tests cleanly before anything built on top of it can run, otherwise you accumulate the learning equivalent of technical debt. Third, real adaptive pacing is a job queue, not a difficulty slider, and it only works if your input signal is honest, which is why school report cards are useless and a separate clean assessment layer is essential. On AI: it is excellent at generating practice problems and explanations, but terrible at judging whether a student genuinely understands, so the component that generates and the component that verifies must never be the same, for the same reason you do not let a class be both writer and auditor. The future of math education is fewer teachers doing higher-leverage work, with the basics offloaded to systems that actually understand mastery instead of performing it.

    9 min

About Society Tech Brief By HackerNoon

From the publisher's feed

Learn the latest society updates in the tech world.