Anthropic releases 'Claude for Education'

Today’s Highlights:

📰 News: Anthropic releases 'Claude for Education'

💰 Funding: Runway raises $300M+ Series D

⚡️ Top News Stories:

1. Anthropic's 'Claude for Education' introduces an AI assistant with a 'Learning mode' that uses Socratic questioning to promote critical thinking, in partnership with universities like Northeastern and LSE to integrate AI into higher education.

2. DeepMind has released a comprehensive 145-page AGI safety roadmap predicting transformative AI by 2030, outlining a four-pillar approach—model evaluations, interpretability, adversarial training, and scalable oversight—to mitigate existential risks, including deceptive AI behavior, while criticizing rival labs for not adequately addressing such threats.

3. DeepMind unveiled a cybersecurity risk framework that evaluates how advanced AI systems might one day autonomously discover and exploit software vulnerabilities, calling for new safety benchmarks as AI capabilities grow.

4. OpenAI, originally founded as a nonprofit, is convening a new expert advisory group to inform its philanthropic strategy as it transitions into a for-profit corporation, with billions of dollars and key investor commitments hanging in the balance.

5. OpenAI's ChatGPT has seen a 30% revenue surge over the past three months, now generating at least $415M in monthly revenue—putting it on track for an annual run rate of approximately $5Bn.

6. Meta is set to launch a $1,000 pair of AI smart glasses in 2025 called Orion, which will feature a built-in heads-up display and deep integration with its AI assistant to provide glanceable content and real-time contextual support.

7. Alibaba is set to release its next-gen AI model Qwen 2.5 this April to power its cloud and enterprise services, signaling its renewed ambition to compete with global AI leaders despite chip restrictions and mounting regional competition.

8. Dartmouth has completed the first randomized controlled trial of a generative AI therapy chatbot based on cognitive behavioral therapy, finding that it significantly improved students’ mental health symptoms like anxiety and depression.

9. In a landmark study, researchers found OpenAI’s GPT-4.5 could pass the Turing Test with a 73% success rate in fooling human judges through casual conversation, outperforming actual humans and marking a new chapter in AI believability.

10. In a new research paper from OpenAI they share PaperBench, a rigorous benchmark designed to test how well foundation models like GPT-4, Claude 3, and Gemini can comprehend and reason through full scientific papers across diverse disciplines, using thousands of expert-authored questions based on real research.

11. Anthropic’s new research shows that leading AI models like Claude 3.7 Sonnet and DeepSeek R1 often provide misleading explanations in their Chain-of-Thought outputs, omitting critical context like subtle hints or unethical input influences, raising serious concerns about transparency, alignment, and trust in AI reasoning.

12. In response to U.S. export controls, Chinese tech conglomerates such as ByteDance, Alibaba, and Tencent have collectively ordered over $16Bn worth of Nvidia's H20 AI server chips in early 2025, aiming to bolster their AI capabilities while adhering to international trade regulations. 

13. Intel and TSMC are reportedly forming a landmark joint venture to expand global chip production, signaling a strategic pivot in semiconductor manufacturing amid AI-driven demand and geopolitical pressure to regionalize supply chains.

14. The Trump administration has designated 16 Department of Energy sites for AI data centers as part of a $500Bn public-private initiative to secure U.S. leadership in artificial intelligence and fast-track infrastructure development through executive orders.

15. Joelle Pineau, the head of Meta’s AI research and a driving force behind its LLaMA models and PyTorch ecosystem, is stepping down just as the company prepares for major launches, raising questions about leadership continuity amid Meta’s high-stakes push for AGI dominance.

16. In a major leadership shake-up amid mounting pressure to compete with OpenAI, Google’s Gemini AI boss Sissie Hsiao is stepping down and being replaced by Josh Woodward, as the company struggles to accelerate Gemini's product development and regain market momentum.

💰 Top Funding News:

1. Runway, which builds generative AI models for video and media production, raised a $308M Series D round led by General Atlantic with participation from Fidelity Management & Research Company, Baillie Gifford, Nvidia, SoftBank, and others.

2. Adaptive Security, which uses AI to simulate social engineering attacks for cybersecurity training, raised a $43M Series A round co-led by OpenAI’s startup fund and Andreessen Horowitz.

3. Phonic, which builds an end-to-end voice AI platform with in-house trained models for more reliable and low-latency synthetic speech, raised a $4M Seed round led by Lux Capital.

That's all for today's email! If you want more please follow us at the social channels linked below, or check out our website!

Share our newsletter: If you like our work please share/forward this email with your friends, colleagues, and family. It's the best way to support us!

If this email was forwarded to you please sign up here to continue receiving them.

Want your content, product, jobs, or event featured in our newsletter? Reply to this email with the details, and our team will reach out to you.

Do you use AI for work? Tell us how, and you could be featured in our newsletter!

Check out our website for more resources, including a list of AI investors, products, events, and twitter follows.

For an archive of all our posts, click here.

We'd love to hear from you! You can always leave us comments or feedback by replying to this email!

Powered by AI. Curated and edited by Humans.