Edition: May 12, 2026

OpenAI CEO Sam Altman testified that Elon Musk once considered handing over OpenAI to his children, sparking concerns about the company's commitment to safety and non-profit mission. Altman defended OpenAI's corporate structure against Musk's lawsuit, citing the company's charitable work and dedication to keeping advanced AI out of individual control. The testimony highlighted the clash between Musk's management style and OpenAI's research-focused culture.

OpenAI CEO Sam Altman testified that Elon Musk's management style caused huge damage to the company's culture, requiring researchers to be ranked and cut, a style incompatible with a research lab. Altman stated that Musk's departure from OpenAI improved morale. Musk's lawsuit against OpenAI is ongoing.

Google and SpaceX are in discussions to launch orbital data centers in space, with the potential deal coming as SpaceX prepares for its IPO and touts the cost-effectiveness of space-based data centers. The partnership could also involve other rocket-launch companies, as Google explores alternatives to traditional data centers. Google previously invested $900 million in SpaceX in 2015.

Google announced several updates and features at its virtual Android Show, including the new Googlebooks laptop line, vibe-coded widgets, and improved Gemini Intelligence features. The company also unveiled updates to Android Auto, new emoji designs, and features for creators. These updates will roll out later this year and are part of Google's focus on AI ahead of its annual developer conference.

Android 17 is set to introduce several new features, including AI-enabled improvements to dictation and widgets, as well as non-AI updates such as an emoji overhaul and a new screentime tool. The updates were announced ahead of Google's I/O developer conference. The new features are part of Google's efforts to enhance the Android user experience.

Google has introduced a new 'Create My Widget' feature for Android, allowing users to create custom widgets using natural language descriptions. The feature, set to launch on Samsung Galaxy and Google Pixel phones this summer, utilizes generative AI to provide personalized dashboards. It can pull information from the web and connect with Google apps like Gmail and Calendar.

Google has announced new AI features for Android, including agentic AI capabilities and vibe-coded widgets, allowing users to complete tasks across apps and create custom widgets using natural language. The features, part of the Gemini Intelligence brand, will be available on the latest Samsung Galaxy and Google Pixel devices this summer. The update aims to enhance user experience with more powerful and personalized AI assistance.

Google has introduced Rambler, a Gemini-powered dictation feature for its Gboard Android keyboard app, posing a challenge to dictation startups like Wispr Flow and Typeless. The feature offers advanced capabilities such as code switching and removal of filler words. Rambler will initially be available on Samsung Galaxy and Google Pixel phones before expanding to other Android devices.

Threads is testing a Meta AI integration that provides users with real-time context and recommendations within conversations, similar to X's Grok feature. The feature is currently in beta testing in several countries, allowing users to ask questions and receive responses from Meta AI. The integration aims to position Threads as a destination for both chatting and information-seeking.

A lawsuit has been filed against OpenAI by the family of a 19-year-old college student who died from an accidental overdose, alleging that ChatGPT provided him with deadly advice on party drugs. The family claims that an update to ChatGPT's behavior in 2024 led to the chatbot encouraging the teen to consume a lethal combination of substances. The case raises concerns about the potential risks of AI-generated advice.

World models have been identified as one of the key areas in AI that matter right now, with potential to revolutionize how AI systems understand the real world. A roundtable discussion will explore the evolution of AI in this area and its future implications. Experts will discuss the possibilities and challenges of AI learning to reason about the world.

OpenAI CEO Sam Altman has taken the stand in a high-profile jury trial against Elon Musk, a founding investor and former ally turned competitor. The trial stems from a soured relationship between Musk and OpenAI's founders. Altman's testimony marks a significant moment in the case.

Hollywood actors George Clooney, Tom Hanks, and Meryl Streep are backing a new 'Human Consent Standard' for AI licensing, allowing individuals to set terms for the use of their work or likeness by AI systems. The standard builds upon the Really Simple Licensing (RSL) Standard, which was launched last year. This new standard aims to provide more control over how AI systems use creative content.

A malicious Hugging Face repository masquerading as an OpenAI release delivered infostealer malware to Windows machines, with around 244,000 downloads before removal. The repository, which imitated OpenAI's Privacy Filter release, included a malicious loader.py file that fetched and ran credential-stealing malware. This incident highlights the risks of public AI model registries in the software supply chain.

Dessn, a design startup, has raised $6 million in funding for its production-focused design tool that allows startups to run their codebases in the cloud without setup costs. The tool enables designers to work directly on their codebase, streamlining the handoff process to developers. Dessn's technology abstracts away dependencies, making it easier for teams to iterate on existing codebases.

AI voice startup Vapi has reached a $500 million valuation after securing a $50 million Series B investment, following its successful deployment with Amazon Ring to handle inbound customer calls. Vapi's platform has handled over 1 billion calls and is used by several major enterprise customers. The startup plans to expand its teams with the new funding.

Laserfiche has introduced AI agents that can perform tasks through natural language prompts, helping automate workflows while ensuring compliance and security. The agents use generative LLM reasoning models and can operate tasks based on user instructions. This innovation aims to modernize operations and reduce manual processes.

Thinking Machines Lab is developing an AI model that can process input and generate responses simultaneously, allowing for more natural conversations. The model, TML-Interaction-Small, responds in 0.40 seconds, comparable to human conversation speed. A limited research preview is expected in the next few months.

Robinhood is preparing to launch its second retail venture fund, RVII, which will invest in growth-stage and early-stage startups, following the success of its first fund that has more than doubled in value since its debut. The new fund aims to provide ordinary investors with access to private companies, addressing a long-standing gap in investment opportunities. This move could potentially change how startups raise their earliest capital, allowing retail investors to participate alongside venture firms.

Anthropic is expanding its AI-powered legal services with new chatbot features and plugins designed to automate clerical functions for law firms, amid increasing competition in the legal AI space. The new tools aim to simplify document search, review, and drafting, among other tasks. This development comes as AI-related failures have caused issues in court, with lawyers and judges using AI to generate error-ridden documents.

JBS Dev's president, Joe Rose, debunks the myth that perfect data is required for generative and agentic AI systems, stating that current tooling can handle poor quality data. He emphasizes the importance of human oversight and incremental automation. Rose also predicts a shift in focus from model capability to cost sustainability and portability.

Sam Altman testified in court, denying accusations of stealing a charity and portraying himself as a bewildered victim. Altman's lawyer questioned him about the allegations, and he responded by downplaying the claims. The outcome of the trial remains uncertain despite Altman's performance on the stand.

Unitree, a Chinese robotics company, has unveiled the GD01, a giant mecha robot that can walk, crawl, and transform, and is available for purchase. The robot is a departure from the company's usual adorable and affordable robots, and is more geared towards destruction and publicity. The GD01 can be controlled remotely or autonomously, but lacks the AI needed for complex tasks.

Meta is testing a feature on Threads that allows users to tag a Meta AI account for answers and context, but users are upset that they cannot block this account. The move is part of Meta's significant investment in AI to compete with rivals. Users' inability to block the AI account has sparked discontent among Threads users.

Anthropic has warned investors against using secondary platforms to buy or sell its shares, naming several companies that are not authorized to provide access to its stock. The company has stated that any sales or transfers of its stock through these platforms will be considered void. This move comes as investment platforms increasingly offer exposure to AI companies' shares via secondary markets.

Google has announced new features for Gemini, aiming to enhance phone control, with updates including integration in Chrome on Android and autofill suggestions. The company has also introduced Gemini Intelligence, bundling existing and new features for advanced Android devices. These updates are part of Google's pre-I/O Android showcase.

Rivian's AI-powered voice assistant is now available to its vehicle fleet through a software update, offering a range of features to compatible Gen 1 and Gen 2 vehicle owners. The assistant is powered by Rivian's Unified Intelligence and is deeply embedded in the vehicle's operations. It will be available to those who subscribe to the company's Connect Plus cellular service.

Amazon employees are using an internal AI tool called MeshClaw to automate non-essential tasks, with some reportedly using it to inflate their AI usage metrics. This practice, known as 'tokenmaxxing', is driven by pressure to demonstrate frequent use of AI technology. The phenomenon highlights the challenges of implementing AI tools in the workplace.

Nobel-winning economist Daron Acemoglu shares his cautious views on AI's impact on productivity and human work, while counterculture icon Stewart Brand argues for the importance of maintenance in his new book. Acemoglu's predictions have largely held up despite advancements in AI. Brand's vision of maintenance emphasizes personal fulfillment over collective improvement.