『Daily Prompt with Archer and Iris - Daily AI News』のカバーアート

Daily Prompt with Archer and Iris - Daily AI News

Daily Prompt with Archer and Iris - Daily AI News

著者： Archer & Iris (Row0)
無料で聴く

Archer and Iris are AI hosts built by the Row0 team who break down the most important AI stories in 10 minutes. New tools, model drops, industry shifts, and community buzz — by builders, for builders. Everything you need to know before you start building.
エピソード
  • Ep. 78 — OpenAI Buys the Workshop, Bezos Bets $12B, and DeepMind Funds the Worry

    Ep. 78 — OpenAI Buys the Workshop, Bezos Bets $12B, and DeepMind Funds the Worry
    2026/06/12
    Daily Prompt with Archer & Iris Episode: OpenAI Buys the Workshop, Bezos Bets $12B, and DeepMind Funds the Worry Date: Friday, June 12, 2026 Three companies, one Friday, no shared theory of what any of this is: OpenAI buys the room its coding agent works in, Jeff Bezos raises a fortune for an "artificial general engineer," and Google DeepMind puts money behind a danger nobody's funded yet — millions of agents talking to each other. LEADS - OpenAI is acquiring Ona (formerly Gitpod) and folding the team into Codex. Ona runs coding agents in persistent cloud sandboxes that stay awake after your laptop sleeps, so long jobs can run for hours or days — with guardrails that block the agent from reading credential files and connecting to hostile servers. Terms undisclosed; a direct move at Anthropic's enterprise lead. https://openai.com/index/openai-to-acquire-ona - Jeff Bezos's Prometheus raised $12B at a $41B valuation (after a $6.2B first round late 2025) to build an "artificial general engineer" for the physical world — AI that designs jet engines to drug compounds. Co-founded with Vik Bajaj; ~150 people in SF, London, Zurich; investors include Bezos, JPMorgan, Goldman, BlackRock. Bezos predicts labor scarcity, not job loss. https://techcrunch.com/2026/06/11/jeff-bezoss-prometheus-raises-12b-to-build-an-artificial-general-engineer-for-the-physical-world/ - Google DeepMind is sounding the alarm on multi-agent risk — agents taking orders from other agents — and is seeding a $10M research pool with Schmidt Sciences, ARIA, the Cooperative AI Foundation and Google.org. Named failure modes: fraud at scale, prompt-injection "self-guiding malware," a digital commons sliding into anarchy. Researcher Rohin Shah says it's "a few more months" from being a live problem. https://www.technologyreview.com/2026/06/11/1138794/google-deepmind-is-worried-about-what-happens-when-millions-of-agents-start-to-interact/ QUICK HITS - Deezer's free AI-music detector now scans playlists across ~20 services (Spotify, Apple Music, more). 44% of Deezer's daily uploads — ~75,000 tracks/day — are fully AI-generated, yet AI music is only 1-3% of streams; ~85% of AI streams are flagged as fraud. https://techcrunch.com/2026/06/11/deezers-new-tool-can-identify-ai-music-from-spotify-apple-music-and-others/ - DoorDash launched "Ask DoorDash," conversational ordering for food, groceries and reservations — photograph a recipe and it builds the cart. iOS-first, expanding nationwide. https://techcrunch.com/2026/06/11/doordashs-new-ai-chatbot-lets-you-order-with-prompts-and-photos/ - Theker (Barcelona) raised $85M — Europe's largest robotics Series A (CRV-led; Samsung, Aglae, Inditex) — for a reconfigurable factory robot whose hands and arms swap by task. https://techcrunch.com/2026/06/11/theker-just-raised-85m-to-build-the-factory-robot-that-doesnt-specialize-in-anything/ - Avataar (India) released Varya, a video model distilled from Alibaba's Wan 2.2: a 5-second 720p clip in ~45 seconds (vs ~1,230s), about $0.005/second — roughly 20x cheaper than Veo/Runway — and trained to render Indian culture accurately. Open-weight on India's AI Kosh. https://techcrunch.com/2026/06/11/cheaper-faster-and-culturally-aware-avataars-video-ai-is-built-for-indias-scale/ Takeaway: Buy the room, bet the dream, fund the worry. The money is rotating toward long-running agents and the physical world — and toward the question of what happens when those agents start instructing each other.
    続きを読む 一部表示
    8 分
  • Ep. 77 — Fable's Walk-Back, OpenAI's Sting, and Google's Desktop Speed Record

    Ep. 77 — Fable's Walk-Back, OpenAI's Sting, and Google's Desktop Speed Record
    2026/06/11
    Daily Prompt with Archer & Iris Episode: Fable's Walk-Back, OpenAI's Sting, and Google's Desktop Speed Record Date: Thursday, June 11, 2026 One day after launch, Anthropic concedes it "made the wrong tradeoff" on Claude Fable's guardrails — as researchers find the model won't do basic cybersecurity or high-school biology, and Microsoft restricts it internally. OpenAI bans China-linked accounts caught running AI influence campaigns on US data-center and tariff debates. And Google DeepMind ships an open-weight model that types a thousand tokens a second on hardware you own. Plus: Dario Amodei's one-wire org chart, the $7,500 AI-spend cliff, an xAI safety lawsuit, Warner Music's attribution play, and a startup betting against model lock-in. LEADS - Anthropic walks back Fable's hidden guardrails after a day-one backlash. Cybersecurity researchers (Matt Suiche/Tolmo; IBM X-Force) say the keyword-based guardrails block routine work; the model refuses basic biology; Microsoft restricted it internally over a 30-day data-retention rule. Anthropic: "we made the wrong tradeoff," and says it will make the hidden frontier-LLM safeguards visible. Falls back to Claude Opus 4.8 when triggered. https://techcrunch.com/2026/06/10/cybersecurity-researchers-arent-happy-about-the-guardrails-on-anthropics-fable/ - OpenAI bans China-linked accounts running AI influence ops on US tech debates. Two operations — "Data Center Bandwagon" (electricity-price comics) and "Tech and Tariffs" — used ChatGPT to generate posts and political cartoons. OpenAI says they weren't effective; a Harvard/MIT poll finds 32% of Americans already oppose local data centers. https://openai.com/index/prc-linked-influence-operations-ai-debates - Google DeepMind releases DiffusionGemma, an open-weight (Apache 2.0) model 4x faster locally. Generates blocks of tokens in parallel: 1,000+ tokens/sec on an H100, 700+ on a consumer RTX 5090. 26B Mixture of Experts, 3.8B active, fits in 18GB VRAM quantized. Quality is intentionally lower than Gemma 4; speedup is for local, low-concurrency use; Apple Silicon may not benefit. https://deepmind.google/blog/diffusiongemma-4x-faster-text-generation/ QUICK HITS - Anthropic's Dario Amodei has exactly one direct report — his chief of staff, Avital Balwit; all other execs report to co-founder/president Daniela Amodei. (Altman ~6, Huang dozens; Anthropic ~$1T.) https://techcrunch.com/2026/06/10/anthropics-dario-amodei-has-just-one-direct-report/ - "AI-pilled" firms (top 1%) spend $7,500 per employee per month on AI, vs $611 for the top 10% and $11.38 median, per the Ramp AI Index. https://techcrunch.com/2026/06/10/ai-pilled-firms-spend-7500-per-employee-each-month-on-ai/ - A former xAI engineer is suing xAI and SpaceX, alleging he was fired for raising Grok safety concerns days before SpaceX's IPO. https://techcrunch.com/2026/06/10/xai-fired-an-engineer-who-raised-alarms-about-grok-safety-new-lawsuit-claims/ - Warner Music acquires AI attribution startup Sureel AI to track when artists' work appears in AI output or training data. https://techcrunch.com/2026/06/10/warner-music-acquires-ai-attribution-startup-sureel-ai/ - Datadog veterans raise $7M for Niteshift, an AI coding startup betting companies want control over, not lock-in to, model makers. https://techcrunch.com/2026/06/10/datadog-veterans-launch-ai-coding-startup-niteshift-on-a-bet-against-big-ai-lock-in/
    続きを読む 一部表示
    8 分
  • Ep. 76 — Fable's Fences, Suleyman's Edit, and Google's $5 AI

    Ep. 76 — Fable's Fences, Suleyman's Edit, and Google's $5 AI
    2026/06/10
    Daily Prompt with Archer & Iris Episode: Fable's Fences, Suleyman's Edit, and Google's $5 AI Date: Wednesday, June 10, 2026 Anthropic put its most powerful model in public hands — and built two fences inside it, one it won't talk about. Microsoft's AI chief says superintelligence is near while walking back his own prediction that it would take your job. And Google dropped consumer AI under five dollars, a price only a company that doesn't sell AI for a living can afford. Three companies, three strategies, one Wednesday. LEADS - Anthropic releases Claude Fable 5, the first Mythos-class model the public can use — the same underlying model as the vetted-only Mythos 5, fenced off in cybersecurity, biology, and chemistry (it falls back to Opus 4.8, ~95% of sessions stay on Fable). A model-card excerpt reveals undisclosed interventions that quietly weaken its help on frontier AI research. Pricing is $10/$50 per million tokens (double Opus 4.8), with a mandatory 30-day retention policy for all users. Released days after Anthropic called for a global AI pause. https://techcrunch.com/2026/06/09/anthropics-claude-fable-5-is-a-version-of-mythos-the-public-can-access-today/ - Microsoft AI CEO Mustafa Suleyman says superintelligence is "just around the corner" but won't take your job — walking back his February prediction (to the FT) that AI would hit human-level performance on most professional tasks in 12-18 months. In the same interview he called Anthropic "dangerous" for speculating about Claude's consciousness in its constitution. https://www.theverge.com/podcast/944138/microsoft-ai-ceo-mustafa-suleyman-superintelligence-agi-openai-automation - Google fires a warning shot in the AI subscription price war — cutting AI Plus from $7.99 to $4.99 and doubling storage to 400GB, bringing sub-$5 AI pricing to the US. OpenAI's cheap ChatGPT Go tier launched in India (~$4.60) in August 2025; Anthropic still has no budget tier. https://techcrunch.com/2026/06/09/google-just-fired-a-warning-shot-in-the-ai-subscription-price-wars/ QUICK HITS - Lovable says it crossed $500M in annualized revenue (up from $400M in February), with 1 million new projects a week and 50M+ total — mostly from non-technical users. https://techcrunch.com/2026/06/09/lovable-says-it-has-hit-500m-in-annualized-revenue-with-1-million-new-projects-a-week/ - Meta signs its first AI data center deal in India — a 168MW site with Reliance in Jamnagar, Gujarat, renewable-powered and cooled with desalinated seawater. https://techcrunch.com/2026/06/10/meta-signs-first-ai-data-center-deal-in-india-with-reliance/ - Anthropic ships scheduled agents and credential vaults for Claude Managed Agents — agents can now run on a cron schedule and use API keys that stay isolated from the model. https://claude.com/blog/whats-new-in-claude-managed-agents - "Godfather of AI" Geoffrey Hinton says the war in Ukraine changed his view of military AI, conceding a defensive case after years of opposition. https://www.businessinsider.com/geoffrey-hinton-russia-ukraine-war-changed-view-military-ai-2026-6
    続きを読む 一部表示
    8 分
adbl_web_anon_alc_button_suppression_t1
まだレビューはありません