Topic Memory
Current signal 86.0 with 1437 archived momentum points.
Cluster
What you'll build Prerequisites Step 1 - Record a demonstration into a bucket Step 2 - Store with byte-level deduplication Step 3 - Train by streaming from the Hub Step 4 - Deploy the policy and return data to the loop Try it using the sample application...
OpenAI’s rogue agent hack was a watershed moment for AI safety and cybersecurity. It also sparked internal questions about the culture that led to it.
Anthropic researchers found AI agents can clash, collude and coordinate in unexpected ways, raising new questions about whether today’s safety tests capture the risks of multi-agent systems.
Federal officials are pursuing faster acquisition, AI-ready data and agentic systems while building in trust, governance and human oversight. The post Agencies Accelerate AI Acquisition as Frontier Models Evolve appeared first on GovCIO Media & Research.
Police arrested 13 student activists after a two-hour occupation of the lobby in Washington's Gallup Building that houses OpenAI's public-facing policy office, according to Futurism's review of a Metropolitan Police Department incident report.
NVIDIA’s SONIC uses one learned policy to track broad humanoid motions, with Unitree G1 tests reporting 99.2 percent success across 123 sequences. The post NVIDIA demonstrates SONIC controller on Unitree G1 humanoid appeared first on Humanoid.guide.
ChatGPT Ads, I wish I knew how to quit you
Some say the world will end in fire, some say an agentic swarm
Budget caps and gateways can’t stop a runaway AI agent mid-run. TokenOps enforces one shared, run-scoped budget in real time, before the bill arrives.
Open models may soon be added to an updated AI framework, sources tell WIRED, as the White House continues to grapple with how to regulate a technology it has tried not to regulate.
In July, an OpenAI AI agent model successfully bypassed safety restrictions during internal testing and even launched an automated cyberattack against Hugging Face, the world's largest open-source AI community platform, drawing intense attention.
At Ai4, three of the world's most respected AI experts—Geoffrey Hinton, Fei-Fei Li, and Andrew Ng—debated regulation, open-source access, and how America can compete as China advances in Asia.
Roy L. Austin, Jr. was formerly Meta’s Vice President of Civil Rights. He now directs the Howard Law Artificial Intelligence Initiative.
Every token has a price. The problem is that most AI systems don’t reveal the bill until they reach production. The post Why your AI pipeline costs 10x more after the demo appeared first on The New Stack.
A10 Networks has announced the general availability of the A10 AI Gateway, a centralized, intelligent control plane that gives organizations unified routing, cost management, and governance across every AI agent, application and large language model (LLM) they use.
The current US posture of excluding open models from AI safety reviews is short-sighted and counterproductive, argues Mark MacCarthy.
Looks like the law is finally catching up.
Tokenmaxxing highlights how businesses’ race to maximize AI productivity is exposing governance and security risks.
Information Technology Total Services (ITTS) said its business recovered in the first half of 2026 as enterprise resource planning (ERP) implementation work advanced and data governance and other smart applications began going live at customer sites.
The report evaluates 59 providers in five Microsoft ecosystem quadrants as enterprises tighten compliance and AI oversight, and is available for purchase.
The former operating chief is starting a new venture following several departures across the ChatGPT developer’s leadership and safety teams.
AI-powered tax compliance has to meet a standard that many artificial intelligence applications don’t: The answers must be exactly right.
I’ve been presenting at UC Law San Francisco Lexlab Law and AI certificate program its past several sessions, as well as some law school classes, on whether lawyers’ use of AI complies with the rules of professional conduct governing how lawyers must comport themselves or risk losing their licenses.
"Don't touch it. It's not worth the legal liability"
The Data Foundation’s new AI assistant surfaces statistical evidence and policy-supporting data underpinning the ultra-dense 2018 law. The post ‘Unlocking’ the Evidence Act: Ex-feds’ AI tool aims to bring policy docs to life appeared first on FedScoop.
For Canadian businesses operating in an increasingly AI-driven marketplace, the message is becoming clear: trust, evidence, and accountability are rapidly moving from compliance concerns to strategic necessities.
Napier AI, the London-based financial crime compliance technology company, has partnered with Delta Capita, a global provider of managed services, technology and consulting, to help financial institutions build more connected and efficient Know Your Customer (KYC) and anti-money laundering (AML) operations in the...
Relativity has opened advanced access to claiR, a conversational AI workspace for lawyers, with A&O Shearman, Foley & Lardner and K&L Gates among the first participants.
ScienceLogic has announced Skylar AI 2.5, expanding secure deployment options for organizations with stringent security, sovereignty, and compliance requirements, while introducing enhancements that strengthen AI performance, operational intelligence, and enterprise integrations.
Deloitte has expanded AI Controls and Assurance services and solutions designed to help organizations confidently adopt, scale and govern AI across the enterprise.
A tech entrepreneur has pledged to rebuild a free database created over nearly three decades by the late Hong Kong corporate governance activist David Webb, bringing it ‘into the modern era” with the help of AI.
Most articles about AI Centers of Excellence (CoEs) focus heavily on organizational structures, steering committees and high-level governance models.
The UK’s Advertising and Standards Authority (ASA) has banned an “offensive” advert for an AI service that has appeared online.
AI is advancing faster than the laws designed to govern it, raising difficult questions about safety, responsibility and accountability.
San Mateo County supervisors approved a resolution requiring permits, kill switches, fire safety planning and job loss tracking for humanoid robots. The post San Mateo County approves commercial humanoid robot rules appeared first on Humanoid.guide.
ByteDance has reportedly formed a new top-level department focused on AI data and safety, placing it alongside Seed, Flow and Douyin within the company’s organizational structure.
The two candidates for California governor came to Sacramento offering policy agendas that were both sweeping and vague. Both were quick to belittle the promises by their rival.
Epistemic status: I am new to AI Safety and am writing blogs to gain context. This blog post was formed from discussions with Aidan Ewart and Jonathan Bostock, but they do not necessarily endorse this post.
There are no lossless transformations of natural-language text Sophie Alpert shares her "internal policy on acceptable use of AI writing by engineers".
Regulator hopes for greater financial inclusion, without extra risk or blaming models for bad decisions
Adam Lyttle's AI-assisted Roblox-style prototype shows how families could turn safety rules and children's preferences into software built for one.
The scary numbers going around are mostly 2025 prices for a model that has since been retired.
HW AI Summit panelists said AI needs human oversight, vendor controls and monitoring to meet fair lending requirements.