Minimalist News

How ChatGPT defends against prompt injection and social engineering by constraining risky actions and protecting sensitive data in agent workflows.

Wayfair uses OpenAI models to improve ecommerce support and product catalog accuracy, automating ticket triage and enhancing millions of product attributes at scale.

How OpenAI built an agent runtime using the Responses API, shell tool, and hosted containers to run secure, scalable agents with files, tools, and state.

IH-Challenge trains models to prioritize trusted instructions, improving instruction hierarchy, safety steerability, and resistance to prompt injection attacks.

ChatGPT introduces interactive visual explanations for math and science, helping students explore formulas, variables, and concepts in real time.

OpenAI is acquiring Promptfoo, an AI security platform that helps enterprises identify and remediate vulnerabilities in AI systems during development.

Codex Security is an AI application security agent that analyzes project context to detect, validate, and patch complex vulnerabilities with higher confidence and less noise.

Descript uses OpenAI models to scale multilingual video dubbing, optimizing translations for both meaning and timing so dubbed speech sounds natural across languages.

See how Balyasny built an AI research system with GPT-5.4, rigorous model evaluation, and agent workflows to transform investment analysis at scale.

OpenAI introduces CoT-Control and finds reasoning models struggle to control their chains of thought, reinforcing monitorability as an AI safety safeguard.

Introducing GPT-5.4, OpenAI’s most most capable and efficient frontier model for professional work, with state-of-the-art coding, computer use, tool search, and 1M-token context.

OpenAI shares new tools, certifications, and measurement resources to help schools and universities close AI capability gaps and expand opportunity.

Practical insights and frameworks to turn AI progress into business advantage

Five AI value models show how leaders can sequence AI from workforce fluency to process reinvention and build durable business advantage.

OpenAI introduces ChatGPT for Excel and new financial app integrations, powered by GPT-5.4 to accelerate modeling, research, and analysis in regulated environments.

A new preprint extends single-minus amplitudes to gravitons, with GPT-5.2 Pro helping derive and verify nonzero graviton tree amplitudes in quantum gravity.

OpenAI introduces the Learning Outcomes Measurement Suite to assess AI’s impact on student learning across diverse educational environments over time.

Axios COO Allison Murphy explains how the company uses AI to support local reporters, streamline newsroom workflows, and deliver high-impact local journalism at scale.

Details on OpenAI’s contract with the Department of War, outlining safety red lines, legal protections, and how AI systems will be deployed in classified environments.

Microsoft and OpenAI continue to work closely across research, engineering, and product development, building on years of deep collaboration and shared success.

Today we’re announcing $110B in new investment at a $730B pre money valuation. This includes $30B from SoftBank, $30B from NVIDIA, and $50B from Amazon.

Stateful Runtime for Agents in Amazon Bedrock brings persistent orchestration, memory, and secure execution to multi-step AI workflows powered by OpenAI.

OpenAI and Amazon announce a strategic partnership bringing OpenAI’s Frontier platform to AWS, expanding AI infrastructure, custom models, and enterprise AI agents.

OpenAI shares updates on its mental health safety work, including parental controls, trusted contacts, improved distress detection, and recent litigation developments.

OpenAI and Pacific Northwest National Laboratory introduce DraftNEPABench, a new benchmark evaluating how AI coding agents can accelerate federal permitting—showing potential to reduce NEPA drafting time by up to 15% and modernize infrastructure reviews.

OpenAI and Figma launch a new Codex integration that connects code and design, enabling teams to move between implementation and the Figma canvas to iterate and ship faster.

Our latest threat report examines how malicious actors combine AI models with websites and social platforms—and what it means for detection and defense.

OpenAI appoints Arvind KC as Chief People Officer to help scale the company, strengthen its culture, and lead how work evolves in the age of AI.

SWE-bench Verified is increasingly contaminated and mismeasures frontier coding progress. Our analysis shows flawed tests and training leakage. We recommend SWE-bench Pro.

OpenAI announces Frontier Alliance Partners to help enterprises move from AI pilots to production with secure, scalable agent deployments.

We share our AI model’s proof attempts for the First Proof math challenge, testing research-grade reasoning on expert-level problems.

OpenAI commits $7.5M to The Alignment Project to fund independent AI alignment research, strengthening global efforts to address AGI safety and security risks.

OpenAI for India expands AI access across the country—building local infrastructure, powering enterprises, and advancing workforce skills.

OpenAI and Paradigm introduce EVMbench, a benchmark evaluating AI agents’ ability to detect, patch, and exploit high-severity smart contract vulnerabilities.

A new preprint shows GPT-5.2 proposing a new formula for a gluon amplitude, later formally proved and verified by OpenAI and academic collaborators.

Introducing Lockdown Mode and Elevated Risk labels in ChatGPT to help organizations defend against prompt injection and AI-driven data exfiltration.

GABRIEL is a new open-source toolkit from OpenAI that uses GPT to turn qualitative text and images into quantitative data, helping social scientists analyze research at scale.

How OpenAI built a real-time access system combining rate limits, usage tracking, and credits to power continuous access to Sora and Codex.

Introducing GPT-5.3-Codex-Spark—our first real-time coding model. 15x faster generation, 128k context, now in research preview for ChatGPT Pro users.

By Ryan Lopopolo, Member of the Technical Staff

OpenAI for Government announces the deployment of a custom ChatGPT on GenAI.mil, bringing secure, safety-forward AI to U.S. defense teams.

OpenAI begins testing ads in ChatGPT to support free access, with clear labeling, answer independence, strong privacy protections, and user control.

AI and cloud technology are reshaping every corner of every industry around the world. Without our customers, there would be no Google Cloud, as they are the ones building the future on our platform. In this regular round-up, we dive into some of the exciting projects redefining businesses, shaping industries, and creating new categories.  For our latest edition, we explore Uber’s revamped network with modern edge; smarter, AI-infused site search at Target; a wicked fast blockchain built by Keeta Network; HCA Healthcare empowers its nurses with a new AI-powered information portal; Deutche Telecom enhances its data science workloads; Google’s AI infrastructure

A year ago today, Google Cloud filed a formal complaint with the European Commission about Microsoft’s anti-competitive cloud licensing practices — specifically those that impose financial penalties on businesses that use Windows Server software on Azure’s biggest competitors.  Despite regulatory scrutiny, it’s clear that Microsoft intends to keep its restrictive licensing policies in place for most cloud customers. In fact, it’s getting worse.  As part of a recent earnings call, Microsoft disclosed that its efforts to force software customers to use Azure are “not anywhere close to the finish line,” and represented one of three pillars “driving [its] growth.” As we approach

AI and cloud technology are reshaping every corner of every industry around the world. Without our customers, there would be no Google Cloud, as they are the ones building the future on our platform. In this regular round-up, we dive into some of the exciting projects redefining businesses, shaping industries, and creating new categories. For our latest edition, we explore Box’s AI agents extracting insights with cross-platform agent integration; Schroders uses multiple connected agents to build a complex investment research system; Hypros’ IoT device can monitor patient distress in hospitals without individual monitoring; a Formula E exhibition of whether regenerative braking

There’s a buzz of excitement here at Tobacco Dock as we welcome our customers and partners to the Google Cloud Summit London. Together, we’re exploring the essential role Google Cloud is playing in driving AI innovation and adoption across the UK. Today is about shining a spotlight on our customers and partners, focusing on the real-world stories of how they are accelerating innovation, realising value, and reshaping entire industries with the powerful combination of AI agents and robust cloud infrastructure. The opportunity ahead is immense. AI is poised to inject more than £400 billion into the UK economy by 2030, and

AI and cloud technology are reshaping every corner of every industry around the world. Without our customers, there would be no Google Cloud, as they are the ones building the future on our platform. In this monthly round-up, we dive into some of the exciting projects redefining businesses, shaping industries, and creating new categories.  For our inaugural edition, we explore Lowe’s new Visual Scout product recommendation engine powered by Vertex AI Vector Search; a responsive “Driver Agent” platform created by Formula E to bring more equity to the track; Ester AI, a conversational tool from wealth.com that can help financial planners

WEEK OF DEC 16 - DEC 20 * Windows Server 2025 is now available on Google Compute Engine. We are excited to announce the general availability of Windows Server 2025 on Google Compute Engine. You can now run Windows Server 2025 Data Center, and Windows Server 2025 Data Center Core editions as well as Windows SQL Server 2022 on Windows Server 2025 with pay-as-you-go licenses. Customers that don’t already have a Microsoft Enterprise Agreement can use the Google Compute Engine’s provided image to take advantage of Google Cloud’s relationships with Microsoft for pay-as-you-go licenses that scale with your workload and

If you’re a regular reader of this blog, you know that 2024 was a busy year for Google Cloud. From AI to Zero Trust, and everything in between, here’s a chronological recap of our top blogs of 2024, according to readership. You’ll probably see some favorites on this list, and discover a few that you missed the first time around.  JANUARY We started the new year strong, removing data transfer fees for anyone moving data off of our platform. Translation: Anyone doing cool things on Google Cloud (like using generative AI to analyze their microservices deployment) is doing it because they want

Over the past year, the pace of innovation across industries and around the world has been extraordinary. At Google, we're inspired by the public sector’s ingenuity and speed in adopting our AI and security innovations to achieve remarkable outcomes together. Since Google Cloud Next, we've launched more powerful versions of Gemini, rolled out Gemini Flash and Imagen 3. And we're proud to showcase inspiring examples from the public sector, highlighting how organizations are leveraging generative AI to drive transformative change. From state agencies streamlining citizen services to healthcare providers improving patient outcomes and defense organizations enhancing mission readiness, public sector organizations

Editors note, November 28th, 2025: “We filed our antitrust complaint with the European Commission (EC) to give voice to our customers and partners about the issue of anticompetitive cloud licensing practices. Today, we are withdrawing it in light of the recent announcement that the EC will assess problematic practices affecting the cloud sector under a separate process. We stand behind the arguments outlined in the blog below, and we continue to work with policymakers, customers, and regulators across the EU, the UK, and elsewhere to advocate for choice and openness in the cloud market." - Giorgia Abeltino, Head of Government

Want to know the latest from Google Cloud? Find it here in one handy location. Check back regularly for our newest updates, announcements, resources, events, learning opportunities, and more.  ---------------------------------------- Tip: Not sure where to find what you’re looking for on the Google Cloud blog? Start here: Google Cloud blog 101: Full list of topics, links, and resources. ---------------------------------------- WEEK OF DEC 10 - DEC 15 * Announcing Launch of Cross Cloud Materialized Views: To help customers on their cross-cloud analytics journey, today we are thrilled to announce the public preview of BigQuery Omni cross-cloud materialized views (aka cross-cloud MVs). Cross-cloud MVs allow customers to very

To our customers: At Google Cloud, we put your interests first. This means that when you choose to work with us, we become partners on a journey of shared innovation, shared support, and shared fate. We are committed to helping you evolve as technology advances, drawing on our depth of experience to ensure you can use the latest and best technology, while keeping you safe and protected. When it comes to the rapidly developing world of generative AI, this is imperative. Earlier this year, we embedded Duet AI — an always-on AI collaborator — across our products, from Google Workspace to Google

Editor’s note: Jack Ngare is a Technical Director in Google Cloud’s Office of the CTO based in Nairobi, Kenya. A native of Kenya, he worked in telecommunications abroad for 12 years before returning home to work for a large African financial institution and a U.S. technology company. He joined Google Cloud last year with a mission to digitize and modernize Cloud’s global top customers, while simultaneously growing Africa's entrepreneurial technology culture.  ---------------------------------------- What is the opportunity you see before you? Working at Google Cloud, I’m able to bring people some of the world's best software and hardware engineering. I also see the possibility

Editor’s note: Alicia Abella is the Managing Director for Telecommunications Industry Solutions at Google Cloud. The work involves bringing complex, advanced technology to businesses across diverse markets and cultures — from startups in Brazil to North America's largest telecommunications companies. The daughter of immigrants from Cuba, she has a PhD, 29 patents, and extensive experience running teams and working with customers. ---------------------------------------- Did you always see yourself bringing new inventions into the world?  No, not at all. I grew up in Queens, New York — the only child of Cuban immigrants. My parents worked hard and saved money so I could get an

We’re excited to share that Gartner has recognized Google as a Leader in the 2023 Gartner® Magic Quadrant™ for Enterprise Conversational AI Platforms, authored by Bern Elliot and Gabriele Rigon. We believe this recognition is a testament to Google Cloud’s robust investments and commitment to innovation in AI, coupled with a deep understanding of enterprise customer needs. Enterprises are increasingly investing in AI-driven solutions that balance addressing customer expectations with operational efficiency. At a time when the demand for quality, performant, and trustworthy conversational AI has never been higher, we’re thrilled to continue to deliver best-in-class technologies, purpose-built to solve our

Editor’s Note: José Neto works in Google Cloud's São Paulo office, assisting customers in both hybrid/on-prem cloud computing systems, and modernizing computer applications for the cloud. It's appropriate work; a native of Salvador, Brazil, he's spent decades harmonizing the past and the future, dealing with radical change, and staying comfortable amid uncertainty as he moves forward.  ---------------------------------------- What do you think defines how you work? It's the way I look at experience. When I work with a customer, I'm interested in helping them fix a problem or build a new service—it's very personal for me. I love working towards results. I always ask what

Welcome back. Anthropic’s newest Claude feature points to one of AI’s most promising uses: helping people understand complex ideas by generating drawings, charts, and other visuals directly inside chat. Microsoft is pushing further into the medical field with Copilot Health, a siloed AI assistant that can connect wearables, health records, and lab results to match you with cited health advice, with strong privacy controls. Meanwhile, Amazon is learning the hard way that AI coding still needs adult supervision, after outages tied to AI-generated code reportedly cost millions in lost orders and triggered new guardrails. AI keeps getting more powerful, but

Welcome back. Nvidia is making one of the smartest moves in AI by open-sourcing not just a powerful new model, but the data, training recipes, and methodology behind it, a strategy that could strengthen the whole ecosystem while driving more demand for Nvidia chips. Anthropic, meanwhile, is sharpening Claude’s value for enterprises with deeper Excel and PowerPoint integrations that carry context across apps and fit nicely into power users' workflows. And Perplexity is pushing agents closer to the mainstream with a Mac mini-based worker that can use local files, coordinate apps, and challenge OpenClaw’s early lead in the race for

Welcome back. World models continue to accelerate as the next frontier in AI. Yann LeCun’s new startup, AMI Labs, just raised a $1 billion seed round to build systems that understand and interact with the real world. And you heard that right, it's a $1B "seed" round. Let that sink in. Google is pushing Gemini deeper into everyday work with new AI features across Docs, Sheets, Slides, and Drive. And Nvidia is placing a strategic bet on Mira Murati’s Thinking Machines, pairing a major investment with plans to deploy massive Vera Rubin systems. Nvidia loves a frontier lab because they

Welcome back. Apple’s new MacBook Pro M5 Max points to a potential reset in AI economics, making it possible to run larger models locally and reduce reliance on costly cloud inference. Microsoft is pushing deeper into agentic work with Copilot Cowork, but the early version looks less capable than more ambitious rival agents. And Anthropic’s clash with the Pentagon is turning into a broader story, as employees from OpenAI and Google publicly back the company’s red lines. —Jason Hiner IN TODAY’S NEWSLETTER 1. RIVAL AI WORKERS LINE UP BEHIND ANTHROPIC 2. MICROSOFT LAUNCHES COWORK AGENT, WITH LIMITS 3. MACBOOK PRO M5 MAX COULD CHANGE AI

Hello, friends. Consumer AI is moving beyond early adopters, and the next breakout AI apps will be powered by personal agents, creative work, and persistent memory. In the latest episode of our podcast, we are joined by a special guest from Andreessen Horowitz to explore the tools gaining traction and the shifts changing how people work. Special thanks to this week’s sponsor, JetBrains, for their support of The Deep View. —Jason Hiner THE CONSUMER AI APPS TO WATCH IN 2026 In the latest episode of The Deep View: Conversations, we talk with Olivia Moore, partner at Andreessen Horowitz (a16z), one of Silicon Valley’s

Welcome back. All three of today's stories show how quickly AI is reshaping the world. At MWC, Qualcomm sketched a future in which AI wearables work together as a body area network rather than a single device. Yann LeCun is challenging the industry’s fixation on AGI, proposing “superhuman adaptable intelligence” as a more practical goal. And in the U.S.–Iran conflict, AI-generated videos are flooding social media, showing how quickly fake videos and images can distort the information landscape. But there's something everyone can do. The technologies advancing AI are powerful; we can all agree on that. The bigger question is

Welcome back. Today, we’re doing a deep dive into Crusoe, an AI factory company that aims to solve the industry’s growing pains at every point in the stack. Special thanks to Crusoe for sponsoring today’s edition of The Deep View. —Jason Hiner IN TODAY’S NEWSLETTER 1. HOW CRUSOE TACKLES AI'S DATA CENTER PROBLEM 2. WHY CRUSOE HITS EVERY PIECE OF THE STACK 3. WHEN MODEL-AGNOSTIC CAN BE THE WINNING STRATEGY  CRUSOE TACKLES AI'S DATA CENTER PROBLEM Infrastructure is perhaps the biggest dilemma facing the AI market today.  Data centers weren’t built with AI in mind. They’re not optimized for high-performance compute and AI workloads. This reckoning has

Hello, friends. Enterprise planning has long been complex, slow, and dependent on specialists. Pigment thinks AI can change that. Its new Modeler Agent lets teams describe planning models in natural language and generate production-ready applications in minutes, aiming to bring the speed of vibe coding to enterprise planning without sacrificing governance, reliability, or scale. This is a special weekend edition of The Deep View, presented in partnership with Pigment. Let us know what you think! —Jason Hiner PIGMENT BRINGS INTENT MODELING TO ENTERPRISE PLANNING Pigment wants to make enterprise planning feel like vibe coding, but production-ready. Enterprise planning software has a dirty secret: most

Welcome back. At MWC, we saw more evidence that the biggest AI opportunities may lie beyond frontier models, with startups using today’s AI to solve concrete problems in medicine, cybersecurity and computer vision. At the same time, Cursor is pushing coding agents into new frontiers beyond just generating code. It's reshaping software development with always-on automation to fill new gaps in the rapidly transforming coding space. And OpenAI’s GPT-5.4 signals that the next battle is for professional workflows, combining stronger reasoning, coding and agentic control into a model built to do more real work with less friction. —Jason Hiner IN TODAY’S

Welcome back. Apple may still look like it’s trailing in the AI race, but its latest research shows it’s quietly working on a critical problem: hallucinations. We break down Apple’s new method for pinpointing exactly where AI goes wrong. At MWC, Honor unveiled a playful “robot phone” with a tiny gimbal arm that hints at how strange AI hardware could soon get. And a new “Pro-Human AI Declaration,” backed by voices from across tech and politics, warns that the race to replace human labor with AI carries momentous societal risks. The pact demands that human thriving gets centered in the

Welcome back. Apple is doubling down on the machines powering the AI boom, with new M5 Pro and M5 Max chips in the MacBook Pro. These machines are built to run large models locally and handle heavier AI workloads for AI builders. At MWC, Alibaba quietly entered the smartglasses race with its Qwen AI glasses, lightweight frames with in-lens displays and swappable batteries — a unique option for glasses. And in the frontier labs race, OpenAI and Google both rolled out faster, lighter models aimed at cutting costs and speeding up everyday AI tasks. Their latest releases show a clear

Welcome back.  Google’s AI glasses are preparing to shake up the AI device market. We got a demo of the latest hardware at Mobile World Congress, and it showed how seamlessly Gemini can blend the digital and physical worlds. Anthropic is courting open-source developers with six months of Claude Max access as an olive branch that’s generous, strategic and notably not open-source. And the biggest story of all: Anthropic’s standoff with the Pentagon has ignited a public debate over who controls frontier AI. The Deep View readers overwhelmingly sided with safety. The power struggle now unfolding will shape contracts, policy,

Welcome back.  AI isn’t killing as many software jobs as we thought. New data shows developer job postings are up 11% year-over-year, as vibe coding lowers the barrier to building custom apps and ultimately drives demand for deeper expertise. Qualcomm just fired the starting gun on the 2026 AI wearables race with its Snapdragon Wear Elite chip, built to upgrade multimodal on-device AI in glasses, pins, earbuds, and more. And in Washington, Anthropic’s refusal to bend on surveillance and autonomous weapons triggered a Pentagon ban, sending its Claude app rocketing to No. 1 on Apple's App Store and intensifying the

Hello, friends. AI is dismantling entry-level jobs and exposing the cracks in resume-based hiring. In this week’s podcast, we examine how AI-powered skills assessments, simulation-based testing, and a free learning platform aim to fix the chaos AI is creating. We also look at why foundational skills, not credentials, will define who thrives in the next era of work. I can’t wait for you to listen to this one because it impacts so many people. Let me know what you think! —Jason Hiner USING AI TO FIX JOB DESTRUCTION, SKILLS, AND HIRING In a labor market being rewired by AI, CodeSignal is betting

Welcome back. If you want a high-paying desk job in 2026, there’s not much hope in avoiding AI. New data from Ladders shows 50% of roles with $100,000-and-above salaries now require AI skills. That's up from just 20% in 2021, signaling that AI literacy has shifted from nice-to-have to core competency. Google’s Nano Banana 2 tackles one of AI image generation’s biggest flaws: text rendering, and potentially unlocking enterprise design use cases in the process. And in Washington, Anthropic is refusing Pentagon pressure to loosen its AI safeguards, betting that long-term trust with enterprises matters more than short-term revenue from