Good morning. It’s Monday, June 16th.

On this day on tech history: In 1977, Larry Ellison, Bob Miner, and Ed Oates established Software Development Laboratories, later Oracle Corporation, which introduced the first widely adopted relational database management system (RDBMS), leveraging Edgar F. Codd’s relational model and Structured Query Language (SQL) to streamline data organization, storage, and access, revolutionizing how businesses manage information.

In today’s email:

• Audio Overviews in Search

• 5 New AI Tools

• Latest AI Research Papers

You read. We listen. Let us know what you think by replying to this email.

Google’s Gemini Expands into Search Audio, Ends Scale AI Alliance

Google is accelerating its multimodal AI agenda. The company is amplifying Gemini’s reach with Audio Overviews in Search, offering hands-free summaries for multitasking users. Available in Labs, the feature uses Gemini’s speech capabilities to deliver compact audio briefings.

Here’s what it looks like:

At the creative frontier, Gemini is also shaping storytelling. ANCESTRA, a DeepMind-produced short film co-created with Darren Aronofsky’s Primordial Soup, premiered at the Tribeca Film Festival. The project combined Veo’s motion-matched video generation, Imagen’s art direction, and Gemini-assisted prompt engineering to blend live-action with AI-generated sequences, rendering complex visuals like a newborn in utero or a journey through the human body.

This momentum comes as Google reportedly exits its $200M contract with Scale AI, following Meta’s $14.3B stake and CEO Alexandr Wang’s move to lead Meta’s superintelligence unit. With Scale’s neutrality in question, Google appears to be doubling down on fully integrated, in-house AI pipelines. Read more.

• New paper pushes back on Apple’s LLM ‘reasoning collapse’ study

• First-of-its-kind technology helps man with ALS ‘speak’ in real time

• Watch: Apple Execs Hint at Foldables, AI Hardware, and iPhone Pricing

• New AI-Accelerated Simulations Reveal Atomic-Level Details of Electrochemical Interfaces

• Anthropic shares blueprint for Claude Research agent using multiple AI agents in parallel

• Tencent releases Hunyuan 3D 2.1 for cinematic asset generation with full open-source support

• ElevenLabs v3 Brings Voice AI to 90% of the World

• Open-source MoE LLM from Rednote challenges top models at a fraction of the cost

• AI models show cognitive structure in triplet odd-one-out test

• GitHub is Leaking Trump’s Plans to 'Accelerate' AI Across Government

• Watch: Nvidia’s Nurabot Rolls Out in Hospitals, Eases Nurse Workload Ahead of Wider Deployment

• IBM aims to build the world’s first large-scale, error-corrected quantum computer by 2028

• Watch: Flowstep Launches AI UI Design Assistant to Turn Prompts into Interfaces and Flows

• Beijing Academy Releases RoboBrain 2.0, Open-Source AI for Perception and Reasoning in Humanoids

• OpenAI is Dorothy: How ‘The Wizard of Oz’ helps explain the dizzying AI landscape

• Sydney team develop AI model to identify thoughts from brainwaves

• Yale students create AI-powered social network

• Rabbit launches "intern," a software AI agent designed to handle team-level projects

• New model helps to figure out which distant planets may host life

📄 ReasonMed: A 370K Multi-Agent Generated Dataset for Advancing Medical Reasoning

📄 SWE-Factory: Your Automated Factory for Issue Resolution Training Data and Evaluation Benchmarks ​​

📄 AniMaker: Automated Multi-Agent Animated Storytelling with MCTS-Driven Clip Generation

📄 Text-Aware Image Restoration with Diffusion Models

📄 Aligned Novel View Image and Geometry Synthesis via Cross-modal Attention Instillation

Thank you for reading today’s edition.

Your feedback is valuable. Respond to this email and tell us how you think we could add more value to this newsletter.

Interested in reaching smart readers like you? To become an AI Breakfast sponsor, reply to this email or DM us on 𝕏!