一种结合了SDE和ODE采样的新型训练框架

IMO2025启示录：多智能体“群体智能”将超越单模？

自动择优的那种

可以期待一下AI的表现

南洋理工西湖数美联合发布

结果对标谷歌和OpenAI，并获陶哲轩认可

8月7日周四，北京线下见～

OpenAI IMO核心团队仅由三名研究人员构成

每月只要1800元

It has been another busy week. GPT-5 appears to be just around the corner… This week, I decode the meaning behind Mark Zuckerberg's "personal superintelligence" manifesto, and what it means for the broader AI race. Keep reading for my chat with a Figma exec on the company's IPO day, a bunch of good ...

Delta Air Lines is explaining more about its AI-assisted dynamic pricing model after coming under scrutiny for recent comments about the pricing. In November, Delta president Glen Hauenstein said at an investor day that “we will have a price that’s available on that flight, on that time, to you, the...

Apple CEO Tim Cook boasted about the potential of AI and the company’s approach to developing it in a rare all-hands today that was reported on by Bloomberg. Apple has been slow to roll out some of its AI features and has stumbled with a planned AI-powered upgrade to Siri, which it delayed earlier t...

Amazon’s new Alexa Plus voice assistant works with most existing Amazon devices. In the week's least surprising news, Amazon CEO Andy Jassy revealed that the company is exploring ways to bring ads to Alexa Plus, its new generative-AI-powered voice assistant. During a conference call following the ...

OpenAI lost access to the Claude API this week after Anthropic claimed the company was violating its terms of service.

On today’s episode of Uncanny Valley, our senior business editor joins us to talk Meta, brain aging, and ChatGPT’s recent dark turn.

On Friday, Anthropic debuted research unpacking how an AI system's "personality" - as in, tone, responses, and overarching motivation - changes and why. Researchers also tracked what makes a model "evil." The Verge spoke with Jack Lindsey, an Anthropic researcher working on interpretability, who ha...

Long before ChatGPT, a group of AI luminaries gathered on an island to discuss the future of artificial intelligence. Their funder ultimately cast a shadow on all who attended.

Microsoft CEO Satya Nadella wanted to make Google dance two years ago, with an AI overhaul of Bing that was designed to steal Google's all-important search market share. Now, there are plenty of signs that Microsoft has successfully taken search share off Google, both in the US and worldwide. Earlie...

Deep Think utilizes extended, parallel thinking and novel reinforcement learning techniques for significantly improved problem-solving.

A “gooner” tells WIRED he became hooked on the cartoonish nature of AI porn. Several addiction experts say the genre could pose a problem for people prone to compulsive sexual behavior.

用5种问题发现AI视频理解的3大弱点

Discover how Figma is transforming digital design with AI. David Kossnick shares how tools like Figma Make empower teams to prototype, collaborate, and build with AI—reshaping workflows for designers, developers, and non-technical creators alike.

Apple CEO Tim Cook says the company is “open” to mergers and acquisitions as it “significantly” increases its investment in AI, as reported by CNBC. Cook’s statements were made alongside the release of Apple’s third-quarter earnings results, which revealed that the iPhone maker raked in $94 billion ...

Microsoft’s stock price has risen so much today that it has passed a $4 trillion market valuation for the first time in its 50-year history. The software maker is the second company to be valued at $4 trillion, after Nvidia reached a market cap of over $4 trillion earlier this month. Microsoft has r...

Behind closed doors, Chinese researchers are laying the groundwork for a new global AI agenda—without input from the US.

Hello, and welcome to Decoder! This is Alex Heath, your Thursday episode guest host and deputy editor at The Verge. Today, I’m joined by Hayden Field, The Verge’s senior AI reporter. We’re talking about the AI talent wars and why some researchers are suddenly getting traded like they’re NBA supersta...

On September 15, WIRED is gathering a panel of leaders across technology, politics, and media to tell you everything you need to know about the future of generative AI.

Personally, I like the white sunglasses look. It was a sweltering Sunday in Brooklyn. Instead of being wasteful and spending $10 on a five-minute Uber, I decided to walk the 20 minutes to a cutesy Japanese stationery cafe with an even cutesier name. Despite the 85 percent humidity, my vanity deman...

We’re launching Stargate Norway—OpenAI’s first AI data center initiative in Europe under our OpenAI for Countries program. Stargate is OpenAI’s overarching infrastructure platform and is a critical part of our long-term vision to deliver the benefits of AI to everyone.

Meta CEO Mark Zuckerberg told investors that his new research lab will focus on building AI models that can learn with minimal human input.

AlphaEarth Foundations is a chip off the Google DeepMind block—and it’s here to help save the world.

RunSybil, a startup founded by OpenAI’s first security researcher, deploys agents that probe websites for vulnerabilities—part of a new AI era for cybersecurity.

Discover how Intercom built a scalable AI platform with 3 key lessons—from evaluations to architecture—to lead the future of customer support.

OpenAI’s new study mode for ChatGPT throws questions back at students, but the learning feature doesn’t address generative AI’s underlying disruption of education.

Introducing study mode in ChatGPT, a new learning experience that helps you work through problems step by step, guiding students with questions, scaffolding, and feedback for deeper learning.

Discover how Outtake uses GPT-4.1 and OpenAI o3 to power AI agents that detect and resolve digital threats 100x faster than before.

Applications are now open for the first-ever Google for Startups Gemini Founders Forum, a two-day, in-person summit hosted by Google DeepMind and Google Cloud.Held Novem…

As part of our Executive Function series, Model ML CEO Chaz Englander discusses how AI-native infrastructure and autonomous agents are transforming financial services workflows.

OpenAI DevDay returns on October 6, 2025 in San Francisco—bringing together 1,500+ developers to preview new tools, hear from OpenAI leaders, and shape the future of AI.

OpenAI and Penda Health debut an AI clinical copilot that cuts diagnostic errors by 16% in real-world use—offering a new path for safe, effective AI in healthcare.

Analysis provides insights into ChatGPT’s impact on the economy. OpenAI also launches new research collaboration to study AI’s broader effects on the labor market and productivity.

Oracle and OpenAI have entered an agreement to develop 4.5 gigawatts of additional Stargate data center capacity in the U.S. This investment will create new jobs, accelerate America’s reindustrialization, and help advance U.S. AI leadership. It also marks a major milestone for Stargate, OpenAI’s AI ...

OpenAI partners with the UK Government to boost AI adoption, drive economic growth, and enhance public services for a thriving AI ecosystem in the UK.

I’ve always considered myself a pragmatic technologist—someone who loves technology not for its own sake, but for the direct impact it can have on people’s lives. That’s what makes this job so exciting, since I believe AI will unlock more opportunities for more people than any other technology in hi...

Gemini Drops is our new monthly update on how to get the most out of the Gemini app.

OpenAI is launching an initial $50 million fund that supports nonprofit and community organizations, informed by the independent OpenAI Nonprofit Commission report.

ChatGPT agent System Card: OpenAI’s agentic model unites research, browser automation, and code tools with safeguards under the Preparedness Framework.

Introducing ChatGPT agent: it thinks and acts, using tools to complete tasks like research, bookings, and slideshows—all with your guidance.

Invideo AI uses OpenAI’s GPT-4.1, gpt-image-1, and text-to-speech models to transform creative ideas into professional videos in minutes.

At OpenAI, we build tools to help people solve hard problems—including nonprofits working on the frontlines of their communities. The OpenAI Academy is teaming up with the Walton Family Foundation, Emerson Collective, and a network of local nonprofit organizations to host the Nonprofit Jam—a one-day...

OpenAI invites researchers to its Bio Bug Bounty. Test the ChatGPT agent’s safety with a universal jailbreak prompt and win up to $25,000.

The Board of Directors thanks the members of the independent OpenAI Nonprofit Commission for their extensive work and engagement.

ChatGPT is designed to be useful, trustworthy, and adaptable—so you can make it your own.

You can now transform your photos into eight-second videos using Veo 3.

Gemini is coming to Wear OS smartwatches — here’s how you can use it.

The latest episode of the Google AI: Release Notes podcast focuses on how Gemini was built from the ground up as a multimodal model — meaning a model that works with tex…

Here are Google’s latest AI updates from June 2025

The latest announcements from Google for Education at ISTE 2025.

Learn more about expanded access to Google Vids for all education users, and Gemini in Classroom, a new suite of no-cost AI tools available for educators.

Learn more about how startups can use Gemini models and other AI resources from Google.

Free and open source, Gemini CLI brings Gemini directly into developers’ terminals — with unmatched access for individuals.

Learn more about AI and inference from Google experts.

The latest episode of the Google AI: Release Notes podcast focuses on how the Gemini team built one of the world’s leading AI coding models.Host Logan Kilpatrick chats w…

Gemini 2.5 Flash and Pro are now generally available, and we’re introducing 2.5 Flash-Lite, our most cost-efficient and fastest 2.5 model yet.

These AI tools from Google I/O 2025 are available globally for your experimentation.

Providing developers with Gemini 2.5 models, personalization capabilities and enhancements to chat.

Extract, built with Gemini, uses the model’s advanced visual reasoning and multi-modal capabilities to help councils turn old planning documents—including blurry maps an…

As we shared at I/O, the Gemini app is getting more personal, proactive and powerful. Starting today, we're rolling out scheduled actions in the Gemini app, a new featur…

Here are Google’s latest AI updates from May 2025

We’re introducing an upgraded preview of Gemini 2.5 Pro, our most intelligent model yet.