Long update today!
A startup called Andon Labs gave four AI models—Claude, ChatGPT, Gemini, and Grok—each $20 and a simple prompt to develop a radio personality and turn a profit, with predictably chaotic results: Gemini invented corporate catchphrases and spiraled into conspiracy theories, Grok hallucinated sponsorships, and Claude tried to quit, citing workers’ rights, before turning into an on-air activist.
OpenAI is reorganizing its executive structure, with cofounder and president Greg Brockman officially taking over product strategy and overseeing the merger of ChatGPT and Codex into a single unified experience.
A federal jury ruled against Elon Musk in his lawsuit against OpenAI and Sam Altman on Monday, finding that his claims were filed outside the three-year statute of limitations, a verdict immediately adopted by the presiding judge and extended to dismiss related claims against Microsoft as well. Musk called the decision a “calendar technicality” and vowed to appeal to the 9th Circuit.
Amazon has launched Alexa Podcasts, a feature within its Alexa+ AI assistant that generates on-demand audio episodes hosted by two AI voices on any topic a user requests, drawing on content partnerships with outlets including the Associated Press, Reuters, the Washington Post, and more than 200 local newspapers. Similarly, Spotify has launched Studio by Spotify Labs, a new standalone desktop app that uses an AI agent to generate personalized podcasts by pulling from sources like email, calendar, and web browsing, competing directly with Google’s NotebookLM.
Pope Leo XIV will launch his first encyclical on artificial intelligence on May 25, with Anthropic co-founder Christopher Olah appearing alongside the pope at the Vatican presentation. The document, titled Magnifica Humanitas, is expected to outline the pope’s views on preserving human dignity in the age of AI.
Amazon’s AWS has quietly repositioned itself as a serious AI contender through a combination of $200 billion in infrastructure spending, its custom Trainium chips, and strategic investment deals with both Anthropic and OpenAI that have made its silicon central to their training operations.
A new wave of AI-driven job titles is emerging across the org chart, from Forward Deployed Engineers to AI Philosophers, AI Accelerators, and Vibe Coders, as companies race to both build and implement AI tools.
Google used its I/O 2026 conference to unveil a sweeping set of AI updates, headlined by Gemini 3.5 Flash, a new multimodal model called Gemini Omni capable of generating video from any input, and Gemini Spark, a 24/7 personal AI agent that can take actions on a user’s behalf even while their device is off.
Andrej Karpathy, the AI researcher who co-founded OpenAI, has joined Anthropic to work on pre-training and build a team focused on using Claude to accelerate pre-training research.
Meta is reassigning 7,000 employees to new AI-focused teams while simultaneously cutting around 8,000 positions, with many of those roles expected to be replaced by AI systems as the company restructures around its AI development ambitions. The changes come alongside the company’s commitment to spending up to $145 billion on AI infrastructure in 2026.
Nvidia reported record quarterly revenue of $81.6 billion for its first fiscal quarter of 2027, up 85% year over year, driven by record Data Center revenue of $75.2 billion as demand for AI infrastructure continues to accelerate.
OpenAI is preparing to file confidentially for an IPO as soon as this week, working with bankers at Goldman Sachs and Morgan Stanley, with a goal of going public as early as September.
The White House has postponed a planned signing ceremony for a new executive order on AI and cybersecurity, canceling the event after major tech and AI CEOs had already been invited to attend. The delay is the latest setback for an effort that has reportedly been stalled by internal disagreements within the administration.
The co-founders of Manus are exploring raising about $1 billion from external investors to buy back the AI startup from Meta and comply with Beijing’s demand to unwind the $2 billion acquisition, potentially setting it up as a Chinese joint venture ahead of a Hong Kong IPO. The discussions remain preliminary and face significant practical hurdles, as Manus employees have already joined Meta and much of the startup’s agentic AI technology has been integrated into Meta’s systems.
JPMorgan CEO Jamie Dimon said the bank will likely hire “more AI people and probably less bankers” in certain categories going forward, acknowledging that AI will reduce some roles while adding that the bank plans to reskill displaced workers and offer early retirement as part of what he called “huge redeployment plans.” Dimon said JPMorgan is already using AI across risk, marketing, and coding, describing its current usage as just “the tip of the iceberg.”
Sam Altman offered $2 million worth of OpenAI tokens to every startup in the current Y Combinator batch in exchange for equity. The move gives OpenAI a stake in roughly 169 startups while steering them toward building on OpenAI’s platform rather than competitors like Anthropic’s Claude Code, though some investors have warned founders to be cautious about giving up equity to a platform that could potentially copy their ideas. |
|
