Howdy! It’s Barsee again.

Happy Thursday, AI family, and welcome back to AI Valley.

Today’s climb through the Valley reveals:

• Meta hires top AI experts from OpenAI for $300 million

• Google’s DeepMind unveils AI for understanding our DNA

• Anthropic now lets you make apps right from its Claude AI chatbot

• Plus trending AI tools, posts, and resources

Let’s dive into the Valley of AI…

TAPLIO

Steal what works. Then make it yours with Taplio

Taplio gives you access to a library of 3M+ viral LinkedIn posts, so you never run out of ideas.

See what’s working in your niche. Get inspired. Then turn that insight into original content — fast.

• Browse top-performing posts by topic, format, or creator

• Use AI to instantly generate a high-quality draft

• Edit, refine, and schedule — all in one place

Try Taplio with the code AIVALLEY1X1 to get 1 month for 1$!

_{*This is sponsored}

THROUGH THE VALLEY

1. Meta hires top AI experts from OpenAI for $300 million

Meta just hired three top AI researchers from OpenAI (Lucas Beyer, Alexander Kolesnikov, and Xiaohua Zhai) by offering them huge deals reportedly worth over $300 million total, including massive $300 million signing bonuses. These researchers had helped set up OpenAI’s Zurich office. Mark Zuckerberg is personally leading Meta’s aggressive recruiting push, sending WhatsApp messages, organizing dinners, and even running a “Recruiting Party” group chat to attract top talent. While some big names like Ilya Sutskever have turned him down, this win shows that Meta’s high-paying offers and personal outreach are starting to work in the AI talent war.

2. Google’s DeepMind unveils AI for understanding our DNA

DeepMind just introduced AlphaGenome, a powerful AI model that can predict how DNA changes affect gene regulation and disease risk. Unlike older models, AlphaGenome can process up to 1 million DNA letters at once, giving high-resolution predictions for thousands of molecular traits like gene expression, RNA splicing, and protein binding. It also outperforms previous tools on almost every benchmark. Available via API for researchers, the model helps explain how even small DNA mutations can cause big biological effects, offering potential breakthroughs in disease research, synthetic biology, and genetics. It’s a major step toward understanding the non-coding 98% of the genome.

3. Federal judge favours Meta in lawsuit over training AI on copyrighted books

A U.S. judge ruled in favor of Meta in a lawsuit brought by 13 authors, including Sarah Silverman, who claimed Meta illegally used their copyrighted books to train its AI. The judge said Meta’s actions qualified as “fair use,” meaning the AI models transformed the content rather than copied it directly, and the authors didn’t show proof of market harm. However, the ruling was narrow, Meta didn’t win on all grounds, and the judge warned future cases with stronger evidence could go the other way. Similar decisions recently favored Anthropic, but lawsuits from media giants like The New York Times and Disney are still pending.

4. Microsoft signals to walk away from negotiations with OpenAI

Tensions are rising between Microsoft and OpenAI as they struggle to renegotiate the terms of their partnership. OpenAI plans to convert into a for-profit company to raise capital and eventually go public, but it needs Microsoft’s approval. Right now, Microsoft holds IP rights to all OpenAI models until 2030 and gets 20% of revenue. OpenAI wants to end that deal by dropping the AGI clause and swapping the revenue share for royalties and equity, but Microsoft isn’t on board. Talks may collapse, with Microsoft signaling it could walk away. If a deal isn’t reached in 6 months, OpenAI faces major setbacks, including losing billions in investment.

5. China builds mosquito-sized spy drone

China’s National University of Defence Technology (NUDT) has developed a mosquito-sized drone for stealth military missions, according to a recent state TV report. The tiny robot, shown held between a student’s fingers, mimics a mosquito and is designed for battlefield reconnaissance and special operations. With delicate wings and thread-thin legs, the drone’s tiny size makes production challenging, requiring advances in materials science, micro-engineering, and bionics. It resembles the U.S.-developed Black Hornet, a palm-sized drone used by global militaries. Similar microdrone research is underway worldwide, including Harvard’s RoboBee and U.S. Air Force projects, with applications also expanding into medicine and environmental monitoring.

PEAK OF THE DAY

Anthropic now lets you make apps right from its Claude AI chatbot

Anthropic announced Wednesday that it will transform its Claude AI assistant into a platform for creating interactive, shareable applications, marking a significant evolution from conversational chatbots toward functional software tools that users can build and distribute without coding knowledge.

Here’s what you need to know:

• Claude AI now allows users to build AI-powered apps directly within the app. This upgrade builds on the Artifacts feature launched last year, which enables users to interact with the creations they ask Claude to make.

• Users can describe the app they want to create, and Claude will write the necessary code. This feature functions like vibe coding, but with the added benefit of being able to see and interact with the results directly inside the Claude interface.

• Early users have built a variety of AI-powered apps, including games, learning tools, data analysis apps, writing assistants, and even agent workflows that orchestrate multiple Claude calls to complete complex tasks.

• The feature is available on Claude’s Free, Pro, and Max tiers, with apps being able to interact with Claude through an API, enabling external integrations and increased functionality.

Why it matters:

The transformation of Claude AI into a platform for creating interactive, shareable applications marks a key shift in AI chatbots. Unlike traditional chatbots, it lets users build apps without coding knowledge, intensifying competition with platforms like OpenAI’s Canvas. The free app creation feature promotes adoption, especially among non-technical users and businesses automating tasks. This move aligns with the rise of citizen development and low-code/no-code tools, making AI development more accessible. Anthropic’s approach blends powerful app creation with ease of use, reshaping the AI development landscape.

TRENDING TOOLS

• Hatch: Work with images, text, and code in one AI canvas (no context lost, just one tool that learns with you). *

• Elevenlabs Voice Design v3: Create any voice you can imagine with a prompt.

• Exa Websets Monitors: Live updates for anything you care about, powered by Exa's neural web index.

• NativeMind: Your fully private, open-source, on-device AI assistant.

• FairPact AI: Scan contracts & find gotchas before signing.

_{*asteric is sponsored}

THINK PIECES / BRAIN BOOST

• Why do some people get incredible AI results while others get garbage? It's because they set a bias with JSON context profiles.

• Begun, the AI browser wars have.

THE VALLEY GEMS

What’s trending on social today:

THAT’S ALL FOR TODAY

Thank you for reading today’s edition. That’s all for today’s issue.

💡 Help me get better and suggest new ideas at [email protected] or @heyBarsee

👍️ New reader? Subscribe here

Thanks for being here.

REACH 100K+ READERS

Acquire new customers and drive revenue by partnering with us

Sponsor AI Valley and reach over 100,000+ entrepreneurs, founders, software engineers, investors, etc.

If you’re interested in sponsoring us, email [email protected] with the subject “AI Valley Ads”.