Kyle Wiggers
- Comms Lead, Ai2
New York
Articles by Kyle Wiggers
Opaque recurrence and other AI terms that you should probably know
; ; ; ; Continue reading More for You Continue reading More for You
Opaque recurrence, and other AI terms that you should probably know
One Royal Will Miss King Harald's Funeral After Being Hospitalized for the Third Time This Year A statement said they're "in good spirits and progressing well under the circumstances." King Harald V of Norway died on August 28 at the age of 89. His state funeral is set to be attended by a plethora of world leaders and royals from across Europe, including Prince William and Princess Anne. …
Opaque recurrence, and other AI terms that you should probably know
AI is rewriting the world and, at the same time, inventing a whole new language to describe how it’s doing it. Sit in on any product meeting, pitch, or panel these days, and you’ll hear people toss around LLMs, RAG, RLHF — and, as of last week, terms like “opaque recurrence,” the reasoning technique in OpenAI’s new Astra model that’s got AI safety researchers rattled. The vocabulary moves fast enough to make even very smart people in the tech world feel a little insecure.
Introducing OlmoEarth embeddings: Custom embedding exports from OlmoEarth Studio for downstream analysis
📄 Tech Report: https://allenai.org/papers/olmoearth | 📊 Documentation: https://docs.olmoearth.allenai.org/embeddings | 💻 Learn more about OlmoEarth: https://allenai.org/olmoearth OlmoEarth Studio, our platform for building Earth observation models, now lets you compute and export embedding vectors—compact numerical representations of Earth-observation data produced by our open source OlmoEarth foundation models.
The only AI glossary you’ll need this year
; ; ; ; Continue reading More for You
The only AI glossary you’ll need this year
Artificial intelligence is rewriting the world and simultaneously inventing a whole new language to describe how it’s doing it. Sit in on any product meeting, pitch, or panel these days, and you’ll hear people toss around LLMs, RAG, RLHF, and a dozen other terms that can make even very smart people in the tech world feel a little insecure.
The only AI glossary you’ll need this year
5 hours ago Artificial intelligence has proven that it can trawl the internet to retrieve information quickly for answering questions. But teaching students using AI is a harder task. The stakes are even higher when the goal is not just learning in school, but performing well on high-stakes exams like the SAT …
So you’ve heard these AI terms and nodded along; let’s fix that
12 hours ago The nationwide Canvas outage that disrupted colleges and universities during finals week forced campuses across the country to improvise as faculty and students suddenly lost access to exams, assignments, grades and other course materials. Canvas is the online learning platform used by over 8,000 …
Everything you need to know about the AI chatbot
ChatGPT, OpenAI’s text-generating AI chatbot, has taken the world by storm since its launch in November 2022. What started as a tool to supercharge productivity through writing essays and code with short text prompts has evolved into a behemoth with 300 million weekly active users.
Everything you need to know about the AI chatbot Original
ChatGPT, OpenAI’s text-generating AI chatbot, has taken the world by storm since its launch in November 2022. What started as a tool to supercharge productivity through writing essays and code with short text prompts has evolved into a behemoth with 300 million weekly active users.
Everything you need to know about the AI chatbot
ChatGPT, OpenAI’s text-generating AI chatbot, has taken the world by storm since its launch in November 2022. What started as a tool to supercharge productivity through writing essays and code with short text prompts has evolved into a behemoth with 300 million weekly active users.
Everything you need to know about the AI chatbot
ChatGPT, OpenAI’s text-generating AI chatbot, has taken the world by storm since its launch in November 2022. What started as a tool to supercharge productivity through writing essays and code with short text prompts has evolved into a behemoth with 300 million weekly active users.
Everything you need to know about the AI chatbot
ChatGPT, OpenAI’s text-generating AI chatbot, has taken the world by storm since its launch in November 2022. What started as a tool to supercharge productivity through writing essays and code with short text prompts has evolved into a behemoth with 300 million weekly active users.
Meta Llama: Everything you need to know about the open generative AI model
Like every Big Tech company these days, Meta has its own flagship generative AI model, called Llama. Llama is somewhat unique among major models in that it’s “open,” meaning developers can download and use it however they please (with certain limitations).
Everything you need to know about the AI chatbot
ChatGPT, OpenAI’s text-generating AI chatbot, has taken the world by storm since its launch in November 2022. What started as a tool to supercharge productivity through writing essays and code with short text prompts has evolved into a behemoth with 300 million weekly active users.
DeepSeek: Everything you need to know about the AI chatbot app Original
DeepSeek has gone viral. Chinese AI lab DeepSeek broke into the mainstream consciousness this week after its chatbot app rose to the top of the Apple App Store charts (and Google Play, as well). DeepSeek’s AI models, which were trained using compute-efficient techniques, have led Wall Street analysts — and technologists — to question whether the U.S. can maintain its lead in the AI race and whether the demand for AI chips will sustain.
Everything you need to know about the AI chatbot
ChatGPT, OpenAI’s text-generating AI chatbot, has taken the world by storm since its launch in November 2022. What started as a tool to supercharge productivity through writing essays and code with short text prompts has evolved into a behemoth with 300 million weekly active users.
Everything you need to know about the AI chatbot
ChatGPT, OpenAI’s text-generating AI chatbot, has taken the world by storm since its launch in November 2022. What started as a tool to supercharge productivity through writing essays and code with short text prompts has evolved into a behemoth with 300 million weekly active users.
Everything you need to know about the AI chatbot
ChatGPT, OpenAI’s text-generating AI chatbot, has taken the world by storm since its launch in November 2022. What started as a tool to supercharge productivity through writing essays and code with short text prompts has evolved into a behemoth with 300 million weekly active users.
Everything you need to know about the AI chatbot
ChatGPT, OpenAI’s text-generating AI chatbot, has taken the world by storm since its launch in November 2022. What started as a tool to supercharge productivity through writing essays and code with short text prompts has evolved into a behemoth with 300 million weekly active users.
Everything you need to know about the AI chatbot
ChatGPT, OpenAI’s text-generating AI chatbot, has taken the world by storm since its launch in November 2022. What started as a tool to supercharge productivity through writing essays and code with short text prompts has evolved into a behemoth with 300 million weekly active users.
Everything you need to know about the AI chatbot
ChatGPT, OpenAI’s text-generating AI chatbot, has taken the world by storm since its launch in November 2022. What started as a tool to supercharge productivity through writing essays and code with short text prompts has evolved into a behemoth with 300 million weekly active users.
Everything you need to know about the AI chatbot
ChatGPT, OpenAI’s text-generating AI chatbot, has taken the world by storm since its launch in November 2022. What started as a tool to supercharge productivity through writing essays and code with short text prompts has evolved into a behemoth with 300 million weekly active users.
Everything you need to know about the AI chatbot
ChatGPT, OpenAI’s text-generating AI chatbot, has taken the world by storm since its launch in November 2022. What started as a tool to supercharge productivity through writing essays and code with short text prompts has evolved into a behemoth with 300 million weekly active users.
OpenAI releases o3-pro, a souped-up version of its o3 AI reasoning model
OpenAI has launched o3-pro, an AI model that the company claims is its most capable yet. O3-pro is a version of OpenAI’s o3, a reasoning model that the startup launched earlier this year. As opposed to conventional AI models, reasoning models work through problems step by step, enabling them to perform more reliably in domains like physics, math, and coding. O3-pro is available for ChatGPT Pro and Team users starting Tuesday, replacing the o1-pro model.
OpenAI releases o3-pro, a souped-up version of its o3 AI reasoning model
1 hour ago This tiny gadget lets me use my AirPods with airplane TVs, and it works flawlessly. If you travel often, you probably already know the struggle: You're settled into your seat, ready to dive into the in-flight entertainment, only to realize your AirPods won't connect to the screen. The airline's …
Atlassian rival Linear raises $82M at $1.25B valuation Original
Linear, an enterprise software maker that competes with many of Atlassian’s products, on Tuesday announced that it raised $82 million in a Series C funding round led by Accel. The round, which also had participation from 01A, Sequoia, Seven Seven Six, and Designer Fund, values Linear at $1.25 billion and brings the company’s total raised to $134.2 million, according to Crunchbase.
Apple's upgraded AI models underwhelm on performance Original
Apple has announced updates to the AI models that power its suite of Apple Intelligence features across iOS, macOS, and more. But according to the company’s own benchmarks, the models underperform older models from rival tech firms, including OpenAI.
Mistral releases a pair of AI reasoning models
French AI lab Mistral is getting into the reasoning AI model game. On Tuesday morning, Mistral announced Magistral, its first family of reasoning models. Like other reasoning models — e.g. OpenAI’s o3 and Google’s Gemini 2.5 Pro — Magistral works through problems step-by-step for improved consistency and reliability across topics such as math and physics. Magistral comes in two flavors: Magistral Small and Magistral Medium.
OpenAI claims to have hit $10B in annual revenue
1 hour ago iPadOS 26 lets Apple’s tablets act more like a Mac than ever before. It also includes a new design that makes all Apple computers look more similar, further bringing its tablets and Macs together. The wraps for iPadOS 26 just came off the update at WWDC, which kicked off on Monday. “iPadOS 26 is our …
Apple brings Apple Intelligence to the iPhone screen Original
Apple said it’s bringing Visual Intelligence, its AI-powered image analysis tech, to the iPhone screen in iOS 26. Visual Intelligence will now make it easier and faster to do more with the content on your iPhone, Apple says, and it works automatically with any app.
Apple lets developers tap into its offline AI models Original
Apple is launching what it calls the Foundation Models framework, which the company says will let developers tap into its AI models in an offline, on-device fashion. Onstage at WWDC 2025 on Monday, Apple VP of software engineering Craig Federighi said that the Foundation Models framework will let apps use on-device AI models created by Apple to drive experiences. These models ship as a part of Apple Intelligence, Apple’s family of models that power a number of iOS features and capabilities.
Corporate AI adoption may be leveling off, according to Ramp data Original
A healthy chunk of corporate America has eagerly embraced AI, betting the tech will bring unrealizable productivity gains. But adoption may be leveling off, according to transaction data from fintech company Ramp. Ramp’s AI Index, which estimates the U.S. business adoption rate of AI products by drawing on Ramp’s card and bill pay data, leveled off at 41% in May after close to 10 straight months of growth.
Anthropic's AI-generated blog dies an early death Original
Claude’s blog is no more. A week after TechCrunch profiled Anthropic’s experiment to task the company’s Claude AI models with writing blog posts, Anthropic wound down the blog and redirected the address to its homepage. Sometime over the weekend, the Claude Explains blog disappeared — along with its initial few posts. A source familiar tells TechCrunch the blog was a “pilot” meant to help Anthropic’s team combine customer requests for explainer-type “tips and tricks” content with marketing goals.
Anthropic’s AI-generated blog dies an early death
11 hours ago Try searching your full name online right now and see what surfaces. Within a few clicks, you’re likely seeing your home address, phone number, age, and even details about family members. If that’s what a simple search turns up, just imagine what malicious actors can find when their business relies …
Qualcomm to acquire semiconductor firm Alphawave Semi for $2.4B
Qualcomm has agreed to acquire Alphawave Semi, a U.K.-based semiconductor company focused on high-speed data center connectivity, for around $2.4 billion. According to Qualcomm CEO Cristiano Amon, the deal will help Qualcomm grow its data center business. “The combined teams share the goal of building advanced technology solutions and enabling next-level connected computing performance across a wide array of high-growth areas, including data center infrastructure,” Amon said in a press release.
OpenAI updates ChatGPT's voice mode with more natural-sounding speech
ChatGPT’s conversational voice mode just got an upgrade. Over the weekend, OpenAI rolled out an update to Advanced Voice, the ChatGPT feature that lets users have back-and-forth dialogues with ChatGPT. Now, ChatGPT’s voices sound more natural and fluid, with “subtler intonation,” “realistic cadence” (including pauses and emphasis), and more “on-point expressiveness” for emotions like empathy and sarcasm, according to the company.
Anthropic appoints a national security expert to its governing trust Original
A day after announcing new AI models designed for U.S. national security applications, Anthropic has appointed a national security expert, Richard Fontaine, to its long-term benefit trust. Anthropic’s long-term benefit trust is a governance mechanism that Anthropic claims helps it promote safety over profit, and which has the power to elect some of the company’s board of directors.
EleutherAI releases massive AI training dataset of licensed and open domain text Original
EleutherAI, an AI research organization, has released what it claims is one of the largest collections of licensed and open-domain text for training AI models. The dataset, called The Common Pile v0.1, took around two years to complete in collaboration with AI startups Poolside, Hugging Face, and others, along with several academic institutions.
OpenAI's marketing head takes leave to undergo breast cancer treatment Original
OpenAI’s head of marketing, Kate Rouch, has announced she’s stepping away from her role for three months while she undergoes treatment for invasive breast cancer. In a post on LinkedIn, Rouch says that Gary Briggs, Meta’s former CMO, will serve as interim head of marketing during her leave. “Earlier this year — just weeks into my dream job — I was diagnosed with invasive breast cancer,” wrote Rouch. “For the past five months, I’ve been undergoing chemo at UCSF while leading our marketing team.
Google says its updated Gemini 2.5 Pro AI model is better at coding Original
Google on Thursday announced an update to its Gemini 2.5 Pro preview model that the company claims is better at certain programming tasks. The company’s calling it an “updated preview,” building on the upgrade to Gemini 2.5 Pro that Google announced around a month ago. Google says the model will roll out in general availability in a “couple of weeks,” and is available starting today in its AI developer platforms AI Studio and Vertex AI and the Gemini app.
Anthropic unveils custom AI models for U.S. national security customers Original
Anthropic says that it has released a new set of AI models tailored for U.S. national security customers. The new models, a custom set of “Claude Gov” models, were “built based on direct feedback from our government customers to address real-world operational needs,” writes Anthropic in a blog post.
North America takes the bulk of AI VC investments, despite tough political environment Original
Despite what some experts have characterized as an environment increasingly hostile to AI R&D, North America continues to receive the bulk of AI venture dollars, according to data from investment tracker PitchBook. Between February and May of this year, VCs poured $69.7 billion into North America-based AI and machine learning startups across 1,528 deals. That’s compared with $6.4 billion that VC firms invested in European AI ventures across 742 deals across the same period.
Hugging Face says its new robotics model is so efficient it can run on a MacBook Original
It’s becoming a little easier to build sophisticated robotics projects at home. AI dev platform Hugging Face released earlier this week an open AI model for robotics called SmolVLA. Trained on “compatibly licensed,” community-shared datasets, SmolVLA outperforms much larger models for robotics in both virtual and real-world environments, Hugging Face claims.
Mistral releases a vibe coding client, Mistral Code
French AI startup Mistral is releasing its own “vibe coding” client, Mistral Code, to compete with incumbents like Windsurf, Anysphere’s Cursor, and GitHub Copilot. Mistral Code, a fork of the open source project Continue, is an AI-powered coding assistant that bundles Mistral’s models, an “in-IDE” assistant, local deployment options, and enterprise tools into a single package. A private beta is available as of Wednesday for JetBrains development platforms and Microsoft’s VS Code.
Mistral releases a vibe coding client, Mistral Code
Kyle Wiggers Wed, 4 June 2025 at 10:00 am GMT-4 2 min read French AI startup Mistral is releasing its own "vibe coding" client, Mistral Code, to compete with incumbents like Windsurf, Anysphere's Cursor, and GitHub Copilot. Mistral Code, a fork of the open source project Continue, is an AI-powered coding assistant that bundles Mistral's models, an "in-IDE" assistant, local deployment options, and enterprise tools into a single package.
Anthropic's AI is writing its own blog — with human oversight | TechCrunch Original
Anthropic has given its AI a blog. A week ago, Anthropic quietly launched Claude Explains, a new page on its website that’s generated mostly by the company’s AI model family, Claude. Populated by posts on technical topics related to various Claude use cases (e.g. “Simplify complex codebases with Claude”), the blog is intended to be a showcase of sorts for Claude’s writing abilities. It’s not clear just how much of Claude’s raw writing is making its way into Claude Explains posts.
DeepSeek may have used Google's Gemini to train its latest model Original
Last week, Chinese lab DeepSeek released an updated version of its R1 reasoning AI model that performs well on a number of math and coding benchmarks. The company didn’t reveal the source of the data it used to train the model, but some AI researchers speculate that at least a portion came from Google’s Gemini family of AI.
AWS establishes new German corporate presence to advance European sovereign cloud Original
Amazon Web Services (AWS), Amazon’s cloud computing division, is forming a new parent company and three subsidiaries in Germany as part of a sovereign cloud the company is launching in the European Union. AWS VP Kathrin Renz will serve as managing director of AWS’ German corporate parent, which will also be led by EU-based “government security and privacy official[s],” according to Amazon.
Salesforce buys Moonhub, a startup building AI tools for hiring Original
Salesforce has quietly snapped up Moonhub, a startup building AI to vet and hire talent. The terms of the deal weren’t disclosed. In a post on Moonhub’s website on Monday, the company said that its entire Menlo Park, California-based team will be joining Salesforce, an early investor in Moonhub. “As a customer and investor, Salesforce has already played a role in Moonhub’s journey,” wrote Moonhub CEO Nanxy Xu in the post.
Show More
loading
Actions
Is this you?
As a journalist, you can create a free Muck Rack account to customize your profile, list your contact preferences, and upload a portfolio of your best work.Get in touch with Kyle
Contact Kyle, search articles and posts on X, monitor coverage, and track replies from one place.
Learn more about Muck Rack