AI updates from Google Cloud Next 25

Today’s Highlights:

📰 News: Google Cloud Next Conference is stacked with AI updates across Google’s suite of products and service 

💰 Funding: Mira Murati aims to raise $2Bn in the largest seed round ever

⚡️ Top News Stories:

1. At Google Cloud Next 2025, Sundar Pichai unveiled major AI upgrades across Google Cloud’s ecosystem, emphasizing how AI is transforming businesses and developers' workflows. Here’s a list of some of the major updates across Google: 

• Pichai highlighted Gemini 1.5 Pro's 1M token context window and announced that Gemini will now be integrated directly into Google Cloud services

• Vertex AI now supports more than 130 foundation models from 100+ model builders, including Anthropic, Mistral, Meta, and Cohere, with users gaining expanded tools to evaluate, tune, and ground these models.

• New innovations include Gemini Code Assist (a competitor to GitHub Copilot), Gemini in Databases for SQL and schema suggestions, and an expansion of AI agent orchestration tools across Vertex AI.

• Google launched the A2A (Agent-to-Agent) protocol to establish a secure, open standard that enables AI agents from different developers to interoperate across platforms, applications, and services.

• Deep Research is available for Gemini Advanced subscribers, enabling users to generate comprehensive, source-backed reports on complex topics using its most advanced AI model, Gemini 2.5 Pro Experimental.

• Google joined OpenAI and others in adopting Anthropic’s Model Context Protocol, a new open standard that enables AI models like Gemini to access and interact with external data sources and applications for more advanced agentic capabilities.

• Ironwood is Google’s most powerful and energy-efficient TPU yet, designed for inference tasks and featuring 4,614 TFLOPs of peak performance, massive memory bandwidth, and optimized support for recommendation workloads.

• Gemini 2.5 Flash, an efficient reasoning model designed for low-latency, high-volume applications like customer support and summarization, offering dynamic compute tuning and soon expanding to on-prem environments.

• Google Workspace is receiving a major AI upgrade with new features like Workspace Flows for automating tasks, podcast-style summaries in Docs, and enhanced tools in Sheets, Meet, and Chat, all while bolstering data compliance through improved residency controls.

• Google expanded its enterprise-focused generative AI suite with Lyria for music generation, Chirp 3 for multilingual voice cloning and transcription, and powerful new editing capabilities in Veo 2 and Imagen 3, all watermarked for traceability but trained on undisclosed data.

• Google’s Gemini Code Assist now features agentic AI tools capable of building apps from specs, rewriting code, and performing reviews and documentation tasks — though human oversight remains key due to ongoing reliability concerns.

• Reddit has enhanced its AI-powered search tool, Reddit Answers, by integrating Google’s Gemini through Vertex AI to improve search accuracy and keep users on-platform with curated comment summaries — a strategic move amid rising AI competition.

• Safe Superintelligence, the new AI startup led by former OpenAI chief scientist Ilya Sutskever, is partnering with Google Cloud to use its TPU chips for research into safe superintelligent AI, signaling a deep infrastructure alignment as it ramps up work toward building advanced frontier models.

• Samsung and Google Cloud are deepening their partnership by integrating Gemini AI models into Ballie, Samsung’s home AI robot, enabling personalized and multimodal assistance through both on-device and cloud-based intelligence.

2. Amazon unveiled Nova Sonic, a next-gen speech foundation model built by the Alexa AI team that delivers faster, more natural, and emotionally aware voice interactions across use cases from storytelling to customer support.

3. OpenAI is launching a new phase of its Pioneers Program to co-create industry-specific benchmarks with select AI startups to better evaluate real-world AI performance in domains like law and healthcare, by empowering these early-stage AI developers with access to new models, funding, and personalized support.

4. OpenAI is preparing to launch several new AI models, including GPT-4.1—a refined version of its multimodal GPT-4o model with integrated reasoning across audio, vision, and text—alongside lighter GPT-4.1 mini and nano variants, and the full release of its advanced o3 reasoning model and a scaled-down o4 mini, all of which have recently appeared in ChatGPT updates.

5. Thinking Machines Lab, the stealthy AI startup founded by ex-OpenAI CTO Mira Murati, continues to add key figures from OpenAI by now enlisting former OpenAI research leaders Bob McGrew and Alec Radford as advisers, signaling serious technical ambition as it works to build customizable, human-aligned AI tools.

6. OpenAI shared an EU Economic Blueprint, arguing that AI could raise Europe’s GDP by up to 20% over the next decade, provided the EU scales sovereign infrastructure, retrains its workforce, and embraces innovation-friendly regulation.

7. Amazon CEO Andy Jassy’s shareholder letter framed AI as a once-in-a-generation opportunity, justifying the company’s $100 billion capital spend in 2025—primarily on AWS data centers and chips—as a critical upfront investment for future dominance.

8. Deep Cogito has launched a family of open-source hybrid AI models that toggle between fast, standard responses and in-depth reasoning, outperforming rival models from Meta and DeepSeek with an ambitious roadmap toward general superintelligence.

💰 Top Funding News:

1. Mira Murati, OpenAI’s former CTO, is reportedly raising a staggering $2Bn for her new startup Thinking Machines Lab—potentially the largest seed round in history—underscoring both investor appetite for generative AI and the rare caliber of her all-star team of ex-OpenAI researchers.

2. NEye Systems, which is building optical switch chips to improve energy efficiency and flexibility in AI data centers, raised a $58M round, led by CapitalG with participation from M12, Micron Technology, and Nvidia.

3. Jobandtalent, which uses AI-powered agents to enhance recruitment and workforce efficiency, raised a $103M Series F, led by Atomico with participation from BlackRock, DN Capital, Hercules, Infravia, Kibo, and Kinnevik.

4. LiveKit, which powers real-time audio and video infrastructure including OpenAI’s Voice Mode and is building scalable services for voice AI agents, raised a $45M Series B, led by Altimeter with participation from Redpoint Ventures and Hanabi Capital.

5. Artisan, which develops AI-powered sales agents to automate outbound outreach and lead management, raised a $25M Series A, led by Glade Brook Capital with participation from Y Combinator, Day One Ventures, HubSpot Ventures, Oliver Jung, Fellows Fund, and others.

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.