🤖 Google brings agentic AI and prompt-built widgets to Android ↗
Google pushed Gemini deeper into Android with agentic features that can move across apps, browse, fill forms, dictate speech, and help build widgets from simple prompts.
The prompt-built widgets part is the sticky bit. You describe what you want, and Android turns it into a homescreen widget. Handy, a little unruly, and very Google.
⚖️ OpenAI chief Sam Altman denies betraying Elon Musk ↗
Sam Altman rejected Elon Musk’s claim that he betrayed OpenAI’s founding mission, saying Musk was the one who wanted control and a path to profit.
The courtroom fight keeps pulling OpenAI’s origin story into public view - idealism, money, power, all tangled like charging cables in a drawer. Nobody comes out looking especially simple here.
🧬 Google-backed Isomorphic raises $2.1 billion to scale AI-driven drug discovery ↗
Isomorphic Labs raised a huge $2.1 billion round to expand its AI drug-design work, with backing from investors including Alphabet-linked funds and others.
The pitch is big: use AI to speed up drug discovery and move candidates toward clinical trials. Big money, big science, and yes, a very expensive bet that biology can be made less stubborn.
🔐 US bank discloses security lapse after sharing customer data with AI app ↗
Community Bank disclosed a security incident tied to an unauthorized AI-based software app, exposing customer information including names, dates of birth, and Social Security numbers.
It is the classic AI-at-work nightmare: someone tries a tool, the data goes somewhere it should never have gone, and suddenly “productivity” feels like a raccoon in the server room.
⚖️ Thomson Reuters and Anthropic Expand Partnership to Connect Claude with CoCounsel Legal ↗
Thomson Reuters and Anthropic expanded their partnership to connect Claude with CoCounsel Legal through an MCP integration.
The goal is to bring Claude into legal workflows while keeping the work inside more controlled, professional-grade systems. Legal AI is getting less “chatbot on the side” and more “quietly inside the machine” - which is probably where the money is.
👀 The Meta AI app now has “live AI.” ↗
Meta added “live AI” to its AI app, letting users point a camera at something and ask questions in real time.
It also added more natural conversation with its Muse Spark model. This is the familiar race now: every assistant wants eyes, ears, and a faintly unsettling sense of presence.
☁️ IBM Announces Red Hat AI Inference and Red Hat OpenShift Virtualization Service on IBM Cloud ↗
IBM announced Red Hat AI Inference on IBM Cloud, aiming to help enterprises deploy AI more securely and predictably.
It is not the flashiest story, but enterprise AI often moves like plumbing, not fireworks. Dull, perhaps. Important, yes - this is where a lot of practical adoption quietly happens.
FAQ
What are the biggest AI tech news stories in this roundup?
The main stories include Google bringing more agentic AI features to Android, OpenAI’s courtroom dispute with Elon Musk, Isomorphic Labs raising $2.1 billion for AI drug discovery, and a US bank disclosing a data exposure tied to an unauthorized AI app. The roundup also covers Claude moving into legal workflows, Meta adding live AI, and IBM expanding enterprise AI services.
How is Google adding agentic AI to Android?
Google is pushing Gemini deeper into Android so it can work across apps, browse, fill out forms, dictate speech, and help users create homescreen widgets from prompts. The prompt-built widget feature lets people describe what they want, then Android turns that request into a usable widget. This points toward Android becoming more task-driven and assistant-led.
Why does the OpenAI and Elon Musk trial matter?
The trial matters because it keeps OpenAI’s founding mission, governance, and commercial direction in public view. Sam Altman denied Elon Musk’s claim that he betrayed OpenAI’s original purpose, arguing instead that Musk wanted control and a path to profit. The dispute highlights the tension between idealistic AI research, money, power, and corporate structure.
What does Isomorphic Labs’ funding mean for AI drug discovery?
Isomorphic Labs raising $2.1 billion signals strong investor interest in using AI to improve drug discovery workflows. The company aims to scale AI-driven drug design and move candidates toward clinical trials. While biology remains complex and unpredictable, the funding shows that major backers are willing to make large bets on AI-assisted science.
What can companies learn from the bank AI data incident?
The bank incident shows why companies need clear rules around employee use of AI apps. Customer information reportedly ended up in an unauthorized AI-based tool, exposing sensitive data such as names, dates of birth, and Social Security numbers. A common safeguard is to restrict unapproved tools, train staff, and keep sensitive workflows inside controlled systems.
How are Claude and CoCounsel being used in legal AI workflows?
Thomson Reuters and Anthropic expanded their partnership to connect Claude with CoCounsel Legal through an MCP integration. The goal is to bring Claude into professional legal workflows while keeping activity inside more controlled systems. This reflects a broader shift from casual chatbot use toward AI tools embedded directly into specialist business software.