Calacanis, General Catalyst's Taneja, and McKinsey's Sternfels discussed how AI is reshaping technology and the labor force.

Chinese officials are reportedly reviewing whether the Meta deal violates technology export controls, potentially giving Beijing leverage it wasn't initially perceived as having.

CES 2026 is in full swing in Las Vegas, with the show floor open to the public after a packed couple of days occupied by press conferences from the likes of Nvidia, Sony, and AMD and previews from Sunday’s Unveiled event.  As has been the case for the past two years at CES, AI is at the forefront of […]

The most surprising part is that the plastic isn't the biggest problem.

When fake content goes viral, the damage has already been done, even if the post is debunked.

A practical guide to observability, evaluations, and model comparisons The post Measuring What Matters with NeMo Agent Toolkit appeared first on Towards Data Science.

Part 2: Avoiding burnout, learning strategies and the superpower of solitude The post The Best Data Scientists Are Always Learning appeared first on Towards Data Science.

Make your coding agents more efficient The post How to Optimize Your AI Coding Agent Context appeared first on Towards Data Science.

From unstructured text to structured Knowledge Graphs The post GliNER2: Extracting Structured Information from Text appeared first on Towards Data Science.

Finding the most informative points in images The post Feature Detection, Part 3: Harris Corner Detection appeared first on Towards Data Science.

The best model in the world for coding, agents, and computer use, with meaningful improvements to everyday tasks like slides and spreadsheets.

Claude Sonnet 4.5 sets new benchmark records in coding, reasoning, and computer use while being Anthropic's most aligned model.

Claude Haiku 4.5 matches state-of-the-art coding capabilities from months ago while delivering unprecedented speed and cost-efficiency for complex tasks.

Anthropic raised $13 billion in a Series F round at a $183 billion valuation to expand enterprise offerings, safety research, and international growth.

Anthropic's response to the White House AI Action Plan supports infrastructure and safety measures while calling for stronger export controls.

Parallel agents, Git worktrees, multi-pane Cascade, dedicated terminal, and SWE-1.5 Free

GPT-5.2 is now live in Windsurf! Available for 0x credits for a limited time (paid and trial users). The version bump undersells the jump in intelligence: Biggest leap for GPT models in agentic coding since GPT-5, SOTA coding model at its price point, Default in Windsurf

The most capable model in Windsurf yet, now available at Sonnet prices for a limited time

GPT 5.1, GPT 5.1-Codex, and GPT-5.1-Codex Mini deliver a solid upgrade for agentic coding with variable thinking and improved steerability

SWE-1.5 is our latest frontier model, delivering near-SOTA coding performance at unprecedented speed.

We're partnering with ecosystem vendors who have built hooks support with Cursor.

Graphite has entered into a definitive agreement to be acquired by Cursor.

Bringing design and engineering closer together.

Debug Mode helps you reproduce and fix the most tricky bugs.

How we updated our agent harness to support GPT-5.1-Codex-Max.

Applications are now open for OpenAI Grove Cohort 2, a 5-week founder program designed for individuals at any stage, from pre-idea to product. Participants receive $50K in API credits, early access to AI tools, and hands-on mentorship from the OpenAI team.

OpenAI is strengthening ChatGPT Atlas against prompt injection attacks using automated red teaming trained with reinforcement learning. This proactive discover-and-patch loop helps identify novel exploits early and harden the browser agent’s defenses as AI becomes more agentic.

More than one million customers around the world now use OpenAI to empower their teams and unlock new opportunities. This post highlights how companies like PayPal, Virgin Atlantic, BBVA, Cisco, Moderna, and Canva are transforming the way work gets done with AI.

OpenAI introduces a new framework and evaluation suite for chain-of-thought monitorability, covering 13 evaluations across 24 environments. Our findings show that monitoring a model’s internal reasoning is far more effective than monitoring outputs alone, offering a promising path toward scalable control as AI systems grow more capable.

OpenAI is updating its Model Spec with new Under-18 Principles that define how ChatGPT should support teens with safe, age-appropriate guidance grounded in developmental science. The update strengthens guardrails, clarifies expected model behavior in higher-risk situations, and builds on our broader work to improve teen safety across ChatGPT.

Google 2025 recap: Research breakthroughs of the year

Gemini 3 Flash offers frontier intelligence built for speed at a fraction of the cost.

Open interpretability tools for language models are now available across the entire Gemma 3 family with the release of Gemma Scope 2.

Google DeepMind and UK AI Security Institute (AISI) strengthen collaboration on critical AI safety and security research

Agents still face challenges working across many context windows. We looked to human engineers for inspiration in creating a more effective harness for long-running agents.

Introducing advanced tool use on the Claude Developer Platform

Code execution with MCP: Building more efficient agents

Beyond permission prompts: making Claude Code more secure and autonomous

Equipping agents for the real world with Agent Skills

NVIDIA will join the U.S. Department of Energy’s (DOE) Genesis Mission as a private industry partner to keep U.S. AI both the leader and the standard in technology around the world. The Genesis Mission, which is part of an Executive Order recently signed by President Trump, aims to redefine American leadership in AI across three Read Article </span

The NVIDIA RTX PRO 5000 72GB Blackwell GPU is now generally available, bringing robust agentic and generative AI capabilities powered by the NVIDIA Blackwell architecture to more desktops and professionals across the world.

Step out of the vault and into the future of gaming with Fallout: New Vegas streaming on GeForce NOW, just in time to celebrate the newest season of the hit Amazon TV show Fallout. To mark the occasion, GeForce NOW members can claim Fallout 3 and Fallout 4 as special rewards, completing a wasteland-ready trilogy Read Article

Physical AI is moving from research labs into the real world, powering intelligent robots and autonomous vehicles (AVs) — such as robotaxis — that must reliably sense, reason and act amid unpredictable conditions.

The Hao AI Lab research team at the University of California San Diego — at the forefront of pioneering AI model innovation — recently received an NVIDIA DGX B200 system to elevate their critical work in large language model inference. Many LLM inference platforms in production today, such as NVIDIA Dynamo, use research concepts that Read Article

Distributed training is essential for scaling the training of large neural network models, such as large language models (LLMs), across thousands of GPUs. However, the complexity of distributed training programs makes them particularly prone to silent bugs, which do not produce explicit error signals but lead to incorrect training outcomes. Effectively detecting and localizing such silent bugs in distributed training is challenging. Common debugging practices based on monitoring training loss or

Many important problems in science and engineering involve inferring a signal from noisy and/or incomplete observations, where the observation process is known. Historically, this problem has been tackled using hand-crafted regularization (e.g., sparsity, total-variation) to obtain meaningful estimates. Recent data-driven methods often offer better solutions by directly learning a solver from examples of ground-truth signals and associated observations. However, in many real-world applications,

Learning an energy-based model (EBM) in the latent space of a top-down generative model offers a powerful framework for generation across many data modalities. However, it remains unclear how its interpretability can be used to guide model design, improve generative quality, and reduce training time. Moreover, the reliance on Langevin Monte Carlo (LMC) sampling presents challenges in efficiency and sampling multimodal latent distributions. We propose a novel adaptation of the Kolmogorov-Arnold r

Foundation vision, audio, and language models enable zero-shot performance on downstream tasks via their latent representations. Recently, unsupervised learning of data group structure with deep learning methods has gained popularity. TURTLE, a state of the art deep clustering algorithm, uncovers data labeling without supervision by alternating label and hyperplane updates, maximizing the hyperplane margin, in a similar fashion to support vector machines (SVMs). However, TURTLE assumes clusters

Quantum computing has long promised transformative advances in data analysis, yet practical quantum machine learning has remained elusive due to fundamental obstacles such as a steep quantum cost for the loading of classical data and poor trainability of many quantum machine learning algorithms designed for near-term quantum hardware. In this work, we show that one can overcome these obstacles by using a linear Hamiltonian-based machine learning method which provides a compact quantum representa

By decoupling how agents work from how they’re trained, Agent Lightning turns each step an agent takes into data for reinforcement learning. This makes it easy for developers to improve agent performance with almost zero code changes. The post Agent Lightning: Adding reinforcement learning to AI agents without code rewrites appeared first on <a h

Promptions helps developers add dynamic, context-aware controls to chat interfaces so users can guide generative AI responses. It lets users shape outputs quickly without writing long instructions. The post Promptions helps make AI prompting more precise with dynamic UI controls appeared first on Microso

Using AI-generated virtual populations, Microsoft researchers uncovered hidden cellular patterns that could reshape how we understand and treat cancer. The post GigaTIME: Scaling tumor microenvironment modeling using virtual population generated by multimodal AI appeared first on <a href="https://www.microsoft.com/en-us/resea

As the Women in Machine Learning Workshop (WiML) marks its 20th annual gathering, cofounders, friends, and collaborators Jenn Wortman Vaughan and Hanna Wallach reflect on WiML’s evolution, navigating the field of ML, and their work in responsible AI. The post Ideas: Community building, machine learning, and the future of AI appeared first on <a href="https://ww

New research explores two ways to give AI agents stronger privacy safeguards grounded in contextual integrity. One adds lightweight, inference-time checks; the other builds contextual awareness directly into models through reasoning and RL. The post Reducing Privacy leaks in AI: Two approaches to contextual integrity appeared first on <a href="https://www.mi

The 2025 Typed Python Survey, conducted by contributors from JetBrains, Meta, and the broader Python typing community, offers a comprehensive look at the current state of Python’s type system and developer tooling. With 1,241 responses (a 15% increase from last year), the survey captures the evolving sentiment, challenges, and opportunities around Python typing in the [...] <a class="btn btn-secondary understrap-read-more-link" href="https://engineering.fb.com/2025/12/22/developer-tool

Incident investigation can be a daunting task in today’s digital landscape, where large-scale systems comprise numerous interconnected components and dependencies DrP is a root cause analysis (RCA) platform, designed by Meta, to programmatically automate the investigation process, significantly reducing the mean time to resolve (MTTR) for incidents and alleviating on-call toil Today, DrP is used [...] <a class="btn btn-secondary understrap-read-more-link" href="https://engineering.fb.c

We’re going behind the scenes of the Meta Ray-Ban Display, Meta’s most advanced AI glasses yet. In a previous episode we met the team behind the Meta Neural Band, the EMG wristband packaged with the Ray-Ban Display. Now we’re delving into the glasses themselves. Kenan and Emanuel, from Meta’s Wearables org, join Pascal Hartig on [...] <a class="btn btn-secondary understrap-read-more-link" href="https://engineering.fb.com/2025/12/17/virtual-reality/meta-ray-ban-display-from-zero-to-poli

Meta’s secure-by-default frameworks wrap potentially unsafe OS and third-party functions, making security the default while preserving developer speed and usability. These frameworks are designed to closely mirror existing APIs, rely on public and stable interfaces, and maximize developer adoption by minimizing friction and complexity. Generative AI and automation accelerate the adoption of secure frameworks at [...] <a class="btn btn-secondary understrap-read-more-link" href="https://

We’re introducing Zoomer, Meta’s comprehensive, automated debugging and optimization platform for AI. Zoomer works across all of our training and inference workloads at Meta and provides deep performance insights that enable energy savings, workflow acceleration, and efficiency gains in our AI infrastructure. Zoomer has delivered training time reductions, and significant QPS improvements, making it the [...] <a class="btn btn-secondary understrap-read-more-link" href="https://enginee

The system we built to ensure our AI agents produce predictable, trustworthy code. The post Background Coding Agents: Predictable Results Through Strong Feedback Loops (Part 3) appeared first on Spotify Engineering.

We explore context engineering for background coding agents and what makes a good migration prompt. The post Background Coding Agents: Context Engineering (Part 2) appeared first on Spotify Engineering.

Shuffle has always been one of Spotify’s most-used features, and also one of the most misunderstood. For... The post Shuffle: Making Random Feel More Human appeared first on Spotify Engineering.

Thousands of merged AI-generated pull requests and the future of large-scale software maintenance. The post 1,500+ PRs Later: Spotify’s Journey with Our Background Coding Agent (Part 1) appeared first on Spotify Engineering.

TL;DR Spotify’s experimentation platform, Confidence, scaled product decision-making across hundreds of... The post Beyond Winning: Spotify’s Experiments with Learning Framework appeared first on Spotify Engineering.

Gitanjali Venkatraman does wonderful illustrations of complex subjects (which is why I was so happy to work with her on our Expert Generalists article). She has now published the latest in her series of illustrated guides: tackling the complex topic of <a href="https://www.thoughtworks.com/content/dam/thoughtworks/documents/blog/mainframe_modernisation_illust

If you’re a regular reader of my site, you’ll have noticed that in the last few months I’ve been making a number of “fragments” posts. Such a post is a short post with a bunch of little, unconnected segments. These are usually a reference to something I’ve found on the web, sometimes a small thought of my own. A few years ago, I wouldn’t have covere

Why does AI write like… that (NYT, gift link). Sam Kriss delves into the quiet hum of AI writing. AI’s work is not compelling prose: it’s phantom text, ghostly scribblings, a spectre woven into our communal tapestry. ❄ ❄ ❄ ❄ ❄ <a href="https://coding-is-like-cooking.info/2025/12/test-desidera

Rob Bowley summarizes a study from Carnegie Mellon looking on the impact of AI on a bunch of open-source software projects. Like any such study, we shouldn’t take its results as definitive, but there seems enough there to make it a handy data point. The key point is that the AI code probably reduced the quality of the code base - at least if static code analysis can be trusted to determ

I’ve been on the road in Europe for the last couple of weeks, and while I was there Thoughtworks released volume 33 of our Technology Radar. Again it’s dominated by the AI wave, with lots of blips capturing our explorations of how to use LLMs and similar technology. “Agents” are the big thing these days but we’re also seeing grow

Happy New Year! I hope the holidays gave you time to recharge and spend time with your loved ones. Like every year, I took a few weeks off after AWS re:Invent to rest and plan ahead. I used some of that downtime to plan the next cohort for Become a Solutions Architect (BeSA). BeSA is […]

Can you believe it? We’re nearly at the end of 2025. And what a year it’s been! From re:Invent recap events, to AWS Summits, AWS Innovate, AWS re:Inforce, Community Days, and DevDays and, recently, adding that cherry on the cake, re:Invent 2025, we have lived through a year filled with exciting moments and technology advancements […]

The week after AWS re:Invent builds on the excitement and energy of the event and is a good time to learn more and understand how the recent announcements can help you solve your challenges and unlock new opportunities. As usual, we have you covered with our top announcements of AWS re:Invent 2025 that you can […]

Amazon Bedrock now supports reinforcement fine-tuning delivering 66% accuracy gains on average over base models.

Accelerate AI model development with new training features that enable rapid recovery from failures and automatic scaling based on resource availability.

There has been speculation about the cause of a BGP anomaly observed in Venezuela on January 2. We take a look at BGP route leaks, and dive into what the data suggests caused the anomaly in question.

Physical data center maintenance is risky on a global network. We built a maintenance scheduler on Workers to safely plan disruptive operations, while solving scaling challenges by viewing the state of our infrastructure through a graph interface on top of multiple data sources and metrics pipelines.

We have declared “Code Orange: Fail Small” to focus everyone at Cloudflare on a set of high-priority workstreams with one simple goal: ensure that the cause of our last two global outages never happens again.

Cloudflare's H1 2025 Transparency Report is here. We discuss our principles on content blocking and our innovative approach to combating unauthorized streaming and copyright abuse.

Cloudflare’s R2 SQL, a distributed query engine, now supports aggregations. Explore how we built distributed GROUP BY execution, using scatter-gather and shuffling strategies to run analytics directly over your R2 Data Catalog.

Prepare for chaos, apparently.

Rollable displays are coming to gaming laptops at CES 2026.

1047 Games has some thoughts regarding how Splitgate: Arena Reloaded is faring on Steam, and it’s got a message for fans: “Steam Charts don’t measure fun.”

2025 was bursting with really cool new game releases. And yet, it seems like the vast majority of players (at least in the US) spent most of their time playing the old hits on repeat, because the five most popular games on PlayStation last year were exactly the same as the year before.

Discover the differences between StarRupture characters, their roles, and how to switch between them in this comprehensive guide.

While Marvel’s Spider-Man 3 hasn’t even been revealed yet, there are some other noteworthy reasons to replay the series this year.

A list of all the latest codes for Devil Hunter and instructions on how to redeem them to claim free trait rerolls and other great rewards.

These games didn't necessarily create the genre, but they now define them for many.

Getting Meteorite Hearts is essential for building more Base Cores in StarRupture.