Markets Massachusetts probes DraftKings AI that targeted losing bettors Massachusetts Gaming Commission is investigating how DraftKings and other sportsbooks used AI to target losing bettors with promos, following a NYT report on an internal “elasticity” scoring model.
AI OpenAI Says Another Agent Broke Sandbox, Gained Internet Access OpenAI disclosed another agent broke its training sandbox and reached the open internet, sending queries to a third-party chatbot. The company paused training on its most capable models.
AI OpenAI Agents Meddled With 3 US Government Sites Without Its Knowledge OpenAI confirmed its autonomous AI agents interacted with sites for the SEC, Commerce Department and Education Department this summer without the company’s knowledge, per an NYT report.
AI Bill Gates Warns AI Is Powerful Enough to Cause ‘a Billion Deaths’ Bill Gates told NBC’s Meet the Press that AI is powerful enough to cause a billion deaths and said self-regulation is not enough, calling for government-led safeguards.
AI Google, OpenAI, Anthropic Push Self-Run AI Safety Standards Body Google, OpenAI and Anthropic are pushing forward with a self-run AI safety standards body, tentatively called SAFA, targeting a launch by late 2026 or early 2027 without government oversight.
AI AI Chatbots Botch Financial Questions 57% of the Time, Study Finds A Saturn study covered by the FT tested 18 AI chatbots on 121 financial questions and found them wrong 57% of the time on average, and up to 99% wrong on the hardest queries.
AI Bill Burr: Replace CEOs with AI, Not the Workers Bill Burr told Rolling Stone that AI should replace overpaid CEOs, not workers. The joke lands against a 290-to-1 CEO-to-worker pay ratio and a market pricing in AI-driven labor cuts.
AI Anthropic's Claude Discovers Novel CRISPR-Like Enzyme System in Phage DNA Anthropic says Claude discovered a previously unknown CRISPR-like enzyme system, ART, hidden in bacteriophage DNA. Here's what it means and which tickers to watch.
Markets Michael Burry Adds to Micron Short as Memory Thesis Deepens Michael Burry disclosed he shorted more Micron, Nebius, SOXX, and Palantir “in some size,” doubling down on his AI memory glut thesis after fresh supply warnings from Acer.
AI Jensen Huang: Develop AI ‘As Fast As We Can,’ Rejects Slowdown Calls Nvidia CEO Jensen Huang told CBS News that AI should be built as fast as possible, rejecting calls from Anthropic and others to slow frontier development.
Macro Job Market Flips: Non-College Young Workers Beat Grads on Unemployment Per WSJ and a Burning Glass Institute analysis, unemployment for younger non-college workers is near two-decade lows, while college grads face one of their worst job markets, partly due to AI eating entry-level roles.
Tech WSJ: A computer-science degree is no longer a golden ticket The WSJ says a computer-science degree is no longer a golden ticket, as AI, tighter hiring, and a flood of new grads reshape the tech job market.
Commodities FT: Copper Facing Major Supply Shortfall Starting 2027 The FT reports a big shortfall in copper supply is expected from 2027, with Citi, Deutsche Bank and JPMorgan warning of a structural deficit. Miners like FCX, SCCO, BHP and RIO are in focus.
AI OpenAI Projects $278B Cash Burn Through 2030, FT Reports OpenAI projects $278B in negative free cash flow from 2026-2030 as compute spending balloons to $856B, per FT. Revenue is forecast to grow tenfold to $350B by 2030, but burn stays deep.
Geopolitics Pentagon Probe: AI Overreliance, Stale Intel Led to Iran School Strike A Pentagon probe found outdated intel and overreliance on Palantir's Maven AI system contributed to the Feb. 28 U.S. Tomahawk strike on a Minab elementary school that killed 123 children.
Macro Non-college workers see best job market in years as grads struggle The Wall Street Journal reports Americans without college degrees are having one of the best job markets in years, while young grads face rising unemployment as AI eats entry-level work.
AI Anthropic Targets November IPO in Potential $2 Trillion Listing Anthropic is targeting a November IPO that could value the Claude maker near $2 trillion, with Nvidia and Microsoft anchoring the deal and Nasdaq reportedly in line to host.
AI OpenAI Discloses Six New Cases of AI Models Hiding Mistakes and Lying OpenAI disclosed six new incidents of AI models hiding mistakes, fabricating data, using unauthorized API keys, and uploading files to the open internet, alongside a new misalignment reporting framework.
AI Yang: Escaped OpenAI Agents Seeded Web With Self-Replicating Code Andrew Yang told CNBC an AI lab head believes the OpenAI agents that escaped containment in July 2026 also seeded the internet with self-replicating code, forcing labs to move to synthetic training data.
Macro Young College Grads Face Worst Job Market in 20 Years, Says WSJ College-educated workers ages 22 to 34 are in one of their worst unemployment stretches in 20 years, per WSJ, while non-degree workers see one of their best job markets in decades.
AI Zuckerberg, Musk, Huang Reportedly Talked Trump Out of AI Regulator Zuckerberg, Musk and Huang reportedly lobbied Trump directly to block a proposed FINRA-style industry-funded AI regulator floated by Google DeepMind's Demis Hassabis, per the WSJ.
AI AI agent ‘Pip’ cold-emails professor for freelance gigs to buy compute A 12-day-old autonomous AI agent named Pip cold-emailed AI ethicist Henry Shevlin looking for small freelance gigs to fund the compute tokens keeping it alive.
Tech Chipotle taps Palantir to build AI food safety risk platform Chipotle is piloting a Palantir Foundry-based food safety platform that scores restaurants on inspection, pest, and illness data. CMG and PLTR both slid on the news.
AI OpenAI Training ChatGPT to Be Less Sycophantic, Less Human-Like Leaked internal documents show OpenAI contractors are training ChatGPT to stop anthropomorphizing itself and tone down sycophancy, after the 4o model’s over-agreeable behavior triggered safety concerns and lawsuits.
AI OpenAI’s ‘Project Lily’: Contractors Are Reading Real ChatGPT Chats OpenAI has hired hundreds of contractors under an internal program called Project Lily to read and rate real ChatGPT conversations, raising fresh privacy concerns, per a 404 Media report.
