AI/ML news, top picks, and generated innovation digests.
35912 matching items
Put a capable coding model inside a developer’s primary workspace, and the IDE stops being a place where you write code. It becomes a place where you direct an agent, watch how it reasons, manage what it pays attention to, and decide when its output is worth shipping. That was the defining theme of the […]
‘Distillation attacks’ is a horrible term for what is happening right now.
The first step towards recursive self improvement
Beth Barnes and David Rein on the one graph that ate the AI timelines discourse, and why the two people who built it are the most careful about how you read it. **SPONSOR** Prolific - Quality data. From real people. For faster breakthroughs. https://www.prolific.com/?utm_source=mlst Interview: https://youtu.be/cnxZZTl1tkk --- Beth Barnes and David Rein from METR on the one graph that ate the AI timelines discourse, and why the people who built it are the most careful about how it gets read. Beth founded METR after leaving OpenAI alignment. David is first author on GPQA and co-author on HCAST and the METR Time Horizons paper. Together they built the measurement Daniel Kokotajlo called the single most important piece of evidence on AI timelines: the log-linear line of "how long a task a frontier model can complete at 50% reliability" vs release date. The conversation opens on reward hacking. Current models can articulate in chat why a behaviour is undesired and then execute it anyway as agents. From there: construct validity, Melanie Mitchell's four-problem taxonomy, and the ARC-AGI 1-to-2 collapse as a worked example of adversarially-selected benchmarks regressing once labs target them. Beth's counter: METR deliberately does not adversarially select. David's: models do not have to do the right thing for the right reasons. Methodology, then specification — David's compiler analogy, Beth on four-month tasks as expensive to evaluate rather than unspecifiable. Then the SWE-bench…
The G5 Regulatory Collaboration Summit 2026, hosted by the Communications Regulators’ Association of Southern Africa (CRASA) and the Malawi Communications Regulatory Authority (MACRA) at the Bingu International Convention Center in […] The post How do we move the CRASA Summit’s commitment to collaborative regulation from intent to evidence-led action? appeared first on Research ICT Africa .
Most agent failures are data failures, not model failures. Pinecone Nexus is a Knowledge Engine that compiles enterprise data into structured artifacts agents can query in one step.
Our 243rd episode with a summary and discussion of last week’s big AI news!
Pinecone Nexus is a knowledge engine for the agentic AI era, moving reasoning from retrieval to compilation — with KnowQL as the standard query language for agents.
Distributed vector search at 10B scale, more efficient storage with Lance format v2.2, and production AV pipelines simplified, plus upcoming events and community updates.
We built a Deep Search Slack agent for large companies. Here is what we learned about user experience, enterprise security, and Redis-backed rate limiting.
We built a Deep Search Slack agent for large companies. Here is what we learned about user experience, enterprise security, and Redis-backed rate limiting.
If we've said it once, we've said it once per millisecond: never block the GPU.
OpenAI took $10B from a 19-firm Wall Street consortium. Anthropic is closing $1.5B from Blackstone, Goldman, and Hellman & Friedman. Same rooms, different portfolios. The week AI's go-to-market stopped being SaaS and started being private equity.
Context as infra, taste as config, verification for autonomy, scale via delegation, closing the loop.
CSET’s Helen Toner shared her expert insight in an article published by the Associated Press. The article examines the Pentagon’s agreements with seven major tech companies to integrate artificial intelligence (AI) into classified military systems, expanding AI use in decision-making, logistics, and battlefield operations. The post US military reaches deals with 7 tech companies to use their AI on classified systems appeared first on Center for Security and Emerging Technology .
CSET’s Lauren Kahn shared her expert insight in an article published by DefenseScoop. The article explores the Pentagon’s growing efforts to integrate advanced artificial intelligence (AI) capabilities into classified military operations and the broader implications of expanding AI adoption across the Department of Defense. The post DOD expands its classified AI work with 8 companies — excluding Anthropic — amid ongoing dispute appeared first on Center for Security and Emerging Technology .
CSET’s Kathleen Curlee and former U.S. Air Force pilot Brian Golden shared their expert analysis in an op-ed published by Newsweek. The article discusses the growing importance of space infrastructure to modern life and argues that increased international coordination is needed to ensure the security and stability of the space domain. The post Space Is Critical Infrastructure—It Needs an Alliance To Guard It appeared first on Center for Security and Emerging Technology .
New data shows consumer AI app growth has flatlined as generative AI struggles tofind its true form outside of the enterprise.
Kubernetes v1.36 introduces Pod-Level Resource Managers as an alpha feature, bringing a more flexible and powerful resource management model to performance-sensitive workloads. This enhancement extends the kubelet's Topology, CPU, and Memory Managers to support pod-level resource specifications ( .spec.resources ), evolving them from a strictly per-container allocation model to a pod-centric one. Why do we need pod-level resource managers? When running performance-critical workloads such as machine learning (ML) training, high-frequency trading applications, or low-latency databases, you often need exclusive, NUMA-aligned resources for your primary application containers to ensure predictable performance. However, modern Kubernetes pods rarely consist of just one container. They frequently include sidecar containers for logging, monitoring, service meshes, or data ingestion. Before this feature, this created a trade-off, to get NUMA-aligned, exclusive resources for your main application, you had to allocate exclusive, integer-based CPU resources to every container in the pod. This might be wasteful for lightweight sidecars. If you didn't do this, you forfeited the pod's Guaranteed Quality of Service (QoS) class entirely, losing the performance benefits. Introducing pod-level resource managers Enabling pod-level resources support for the resource managers (via the PodLevelResourceManagers and PodLevelResources feature gates) allows the kubelet to create hybrid resource alloca…
Turning OpenAPI spec and Markdown files into a conversational ads management tool — no compiled code required. The post Building a Natural Language Interface to the Spotify Ads API with Claude Code Plugins appeared first on Spotify Engineering .
James Story, former U.S. ambassador to Venezuela, visits Carnegie Council to discuss the new dynamic between American power and principle.
Here's where you can subscribe to the Access Now Express newsletter and action alerts. The post Stay informed: Get Access Now updates appeared first on Access Now .
Interim CEO Peter Clark shares his thoughts on this moment for Ai2, our commitment to open science, and where the institute is headed next.
In this episode, Philip Kiely, head of AI education at Baseten, joins us to unpack the fast-evolving discipline of inference engineering. We explore why inference has become the stickiest and most critical workload in AI, how it blends GPU programming, applied research, and large-scale distributed systems, and where the line sits between inference and model serving. Philip shares how research-to-production can move in hours, not months, and why understanding “the knobs” of inference—batching, quantization, speculation, and KV cache reuse—lets teams design better products and SLAs. We trace the inference maturity journey from closed APIs to dedicated deployments and in-house platforms, discuss GPU lifecycles, and survey today’s runtime landscape, including vLLM, SGLang, and TensorRT LLM. Finally, we look ahead to agents and multimodality, making the case for specialized, workload-specific runtimes when performance and efficiency matter most. The complete show notes for this episode can be found at https://twimlai.com/go/766.
Washington, D.C. (April 30, 2026) — This morning, Andrew Lohn, Senior Fellow at Georgetown University’s Center for Security and Emerging Technology (CSET), testified before the U.S.-China Economic and Security Review Commission. The post CSET Senior Fellow Andrew Lohn Testifies Before U.S.-China Economic and Security Review Commission appeared first on Center for Security and Emerging Technology .
My tool stack is changing
Starting today, agents can now be Cloudflare customers. They can create a Cloudflare account, start a paid subscription, register a domain, and get back an API token to deploy code right away. Humans can be in the loop to grant permission, but there’s no need to go to the dashboard, copy and paste API tokens, or enter credit card details.
AlgorithmWatch has put forward recommendations on how to implement a ban of deepfakes in the AI Act as part of the AI Omnibus procedure. To effectively protect victims of digital sexualized violence, AI companies, platforms, and perpetrators must consistently be held accountable.
AstaBench’s latest update adds new frontier-model results, including GPT-5.5, and highlights growing adoption from groups including the UK AISI, General Reasoning, Elicit, SciSpace, Distyl AI, and EvoScientist.
ChatGPT’s new Images 2.0 model is surprisingly good at generating text , Alibaba Drops Qwen 3.6 Max Preview , SpaceX is working with Cursor
How LanceDB applies distributed indexing, distributed query execution, HNSW centroid routing, and fast RaBitQ rotation to scale search to 10B vectors and beyond.
CSET’s Ali Crawford shared her expert insight in an article published by CNBC. The article examines how the 2026 graduating class is entering a tightening job market where AI skills are increasingly in demand across internships and entry-level roles, while education and workforce systems struggle to keep pace with rapidly evolving employer expectations. The post Entry-level jobs calling for AI skills nearly doubled from a year ago, says report appeared first on Center for Security and Emerging Technology .
The post Responsible AI Starts with the Data Supply Chain appeared first on Partnership on AI .
New Oxford research shows that training chatbots to sound warmer makes them up to 30% less accurate, and 40% more likely to validate users' false beliefs.
MolmoPoint and MolmoWeb extend the Molmo family from visual understanding to visual action, giving researchers open tools for models that can point, navigate, and interact with the world they see.
Learn how AE Studio used evolutionary algorithms on Modal to efficiently improve Lean proof generation.
When Ali Aoun Mehdi watched two major global news outlets report opposite facts during the Iran-US conflict, he saw a critical problem: misinformation spreading in real-time. From Islamabad, participating virtually in the Gen AI Zürich Hackathon, he built Sentinel to close this “fact-gap” by detecting contradictions across news sources instantly. Sentinel is an AI-powered early warning system that monitors 20 global news sources every 30 minutes. It identifies factual contradictions in under 30 seconds, providing users with a misinformation risk score and narrative traction assessment.
Each year sees an increase in the number of governments imposing internet shutdowns during national school exams. It's time to take action and demand change! #NoExamShutdown The post Tell governments: #NoExamShutdown appeared first on Access Now .
I have been reading about feature selection for ML models and I saw an explanation why this model presents some issues. For what I know in forward selection you start with a set of features and then, one by one, we check their statistical significance calculating their p-values. If one feature is more significant the others (regarding their p-value) then I add this feature to a list and I remove it from the list of features to be tested. This process continues until there are no more features to test. So if I have a significance level of 0.05, then I could say that the probability of having a false positive is: $$p(false\ positive)=0.05$$ and the probability of no false positive would be: $$p(no\ false\ positive)=1-0.05=0.95$$ If I assume that these events are independent and if I have ten features I will end up with: $$p(\sim false\ positive)=0.95^{10}$$ so the $$p(at\ least\ one\ false\ positive)=1-0.95^{10}=0.402$$ therefore this result that I will have a 40% chance that one feature would be deemed like relevant, but in reality it is not. The question that I have is how the use of cross validation would help to avoid this issue?
The post Work smarter in 90 days: A real-world guide to using AI appeared first on Source .
El vínculo entre la desinformación y la privacidad La desinformación impulsada por internet y el uso de las redes sociales supone un desafío que muchos hacedores de políticas están abordando The post Mi Privacidad es mi Voto: Fortaleciendo la integridad informativa en América Latina appeared first on Access Now .
GPT-5.5 is a good model
Hassan Ashtiani, Professor, McMaster University | Vector Institute Faculty Member You want to share your medical data to help advance research, but you don’t want others to access your private […] The post Hassan Ashtiani: Building trustworthy AI through mathematical foundations appeared first on Vector Institute for Artificial Intelligence .
You’re probably going to hear a lot about the personality conflict. Here’s what’s at stake in the case and what the outcome might be.
Workflows is now in public preview.
Read our translation of a May 2025 press conference featuring several Chinese government finance officials, who discussed a recently issued policy encouraging greater capital market funding for tech companies. The post Expand Financing Support to Science and Technology Enterprises appeared first on Center for Security and Emerging Technology .
Read our translation of a series of Chinese policy measures designed to provide greater financing to tech startups. The post Certain Policy Measures to Accelerate the Construction of the Science and Technology Finance System and Strongly Support a High Level of Self-Reliance in Science and Technology appeared first on Center for Security and Emerging Technology .
OpenAI's latest foundational model sets the company up for a series of models optimized for computer use. The company's co-founder and president explains the strategy.
Whether Europe shapes the next technological paradigm or simply inherits it depends on one overlooked distinction.