OpenAI + Google submit policy recs to The White House

Today’s Highlights:

📰 News: OpenAI and Google submit policy recommendations to The White House for upcoming U.S. AI Action Plan + OpenAI releases new tools for creating AI agents + Google launches Robotics model

💰 Funding: Google to inject $750M into Anthropic

⚡️ Top News Stories:

1. OpenAI has submitted recommendations to the White House Office of Science and Technology Policy (OSTP) for the upcoming U.S. AI Action Plan

• The comprehensive set of recommendations to the White House focus on regulatory, export control, and copyright strategies to bolster America's AI leadership, promote innovation, and safeguard national security.

• In particular it has escalated its opposition to Chinese AI lab DeepSeek, labeling it “state-controlled” and urging U.S. policymakers to ban PRC-backed models due to security and IP theft concerns.

• They have also urged the U.S. government to adopt a copyright strategy that permits AI models to continue learning from copyrighted material, arguing that fair use has been instrumental in America's AI dominance by fostering investment, research, and innovation.

2. Google has also submitted its AI policy proposal advocating for weaker copyright restrictions and "balanced" export controls, emphasizing that broad fair use and text-and-data mining exceptions are essential for maintaining U.S. AI innovation and global competitiveness.

3. OpenAI launches Responses API and Agents SDK to equip developers with powerful tools to create AI agents that can browse the web, retrieve files, and automate workflows, while addressing the technical challenges that still limit agent reliability.

4. Google DeepMind has introduced Gemini Robotics, an AI model that enables robots to interact with objects, navigate environments, and perform real-world tasks with greater adaptability, demonstrated through complex actions like folding paper and placing glasses into a case, with a variant, Gemini Robotics-ER, specializing in embodied reasoning for efficient task execution.

5. Google releases Gemma 3, an open-source AI model optimized for single GPU or TPU usage, offers developers enhanced multimodal analysis capabilities and supports over 35 languages, positioning it as a high-performance alternative to existing models.

6. Google has expanded Gemini 2.0 Flash's native image generation capabilities, enabling multimodal input and enhanced text rendering for creative applications. 

7. OpenAI’s Chain-of-Thought Monitoring introduces a new method to improve AI safety by enabling models to explain their reasoning processes, making it easier to detect misaligned behaviors like reward hacking.

8. OpenAI CEO Sam Altman revealed that the company has trained a new AI model specializing in creative writing, showcasing a metafictional short story that he described as "strikingly good," though OpenAI has yet to decide on its release.

9. Japanese AI startup Sakana claims its AI, The AI Scientist-v2, successfully generated research papers end-to-end—including hypotheses, experiments, data analyses, and visualizations—with at least one passing peer review.

10. China’s Manus AI has partnered with Alibaba’s Qwen AI team to accelerate its development of a general AI agent, which it claims outperforms OpenAI’s DeepResearch.

11. Meta is testing its first in-house AI training chip, part of its Meta Training and Inference Accelerator (MTIA) series, aiming to reduce reliance on Nvidia and lower infrastructure costs, with initial focus on recommendation systems before expanding to generative AI applications like Meta AI.

💰 Top Funding News:

1. ServiceNow is acquiring AI startup Moveworks for $2.85Bn, its largest acquisition yet, to integrate Moveworks’ AI assistants into its platform, enhancing AI-driven automation and self-service tools with minimal human supervision.

2. Google owns a 14% stake in Anthropic and has invested over $3Bn in the AI startup, with an additional $750M set to be injected in 2025 through a convertible debt deal, according to new court filings.

3. Salesforce is investing $1Bn in Singapore over the next five years to drive the adoption of Agentforce, its AI-powered autonomous agent development platform, aiming to enhance workforce efficiency and strengthen its presence in Southeast Asia.

4. Cartesia, specializing in ultra-realistic, low-latency voice AI models like Sonic 2.0, raised a $64M Series A led by Kleiner Perkins, with participation from Index Ventures, Lightspeed, A\*Capital, Factory, Greycroft, Dell Technologies Capital, and Samsung Ventures. 

5. Bria, an AI-powered image generation startup, raised a $40 million Series B round led by Red Dot Capital, with participation from Maor Investments, Entrée Capital, GFT Ventures, Intel Capital, and IN Venture, bringing its total funding to $65 million.

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.