AI News Today — Part 2
Sep 5, 2026 · Sourced from 500+ daily AI sources, scored by relevance.
- When AI Takes Notes: Court Allows Privacy Claims Against Otter.ai to Proceed
Otter.ai's AI notetaker tool has been accused of violating Illinois's biometric privacy law, and the California court hearing the case recently let the proceedings move forward. As we have previously written in our sister blog, AI notetaker ... (https://incidentdatabase.ai/cite/1650#7893)
- Court Filings In A.I. Suit Invoke Copyright Law, Culture and Sports
Filings made Friday in The New York Times’s closely watched lawsuit against OpenAI and Microsoft included a range of copyright law and cultural references.
Score: 48🌐 MovesSep 5, 2026https://www.nytimes.com/2026/09/04/technology/openai-microsoft-new-york-times-lawsuit.html - NVIDIA Releases Personal AI Router (PAIR): An Open Source Virtual Inference Router that Distributes Local AI Requests Across RTX, DGX Spark, and Mac Nodes
NVIDIA Releases Personal AI Router (PAIR): An Open Source Virtual Inference Router that Distributes Local AI Requests Across RTX, DGX Spark, and Mac Nodes MarkTechPost
- Chrome could soon let you ask Google about anything on your screen — even outside the browser
Chrome could soon let you ask Google about anything on your screen — even outside the browser Tom's Guide
- Minnesota nudification law survives xAIs request for a legal injunction
Minnesota has temporarily won the right to sue companies for up to $500,000 for every "nudification" enabled by artificial intelligence.
Score: 47🌐 MovesSep 5, 2026https://mashable.com/tech/minnesota-nudification-law-survives-xais-request-for-a-legal-injunction - ASEAN can still hedge between America and China on AI. It needs to get its act together first
ASEAN can still hedge between America and China on AI. It needs to get its act together first Fortune
- Researchers report on a solar storm that caused a GPS glitch serious enough to crash self-driving cars
The solar storm that hit Earth last November was more severe than most, and we need to be prepared for the next one.
- ‘We’re plausibly close to crossing the line’: are warnings of uncontrollable AI coming true?
A spate of serious safety incidents have increased fears about the power and impenetrability of the most advanced models Picture humanity in a boat being swept down a raging river, praying there is no Niagara Falls ahead. Or imagine standing with the pioneering physicists in 1942 before they triggered the first self-sustaining nuclear fission chain reaction beneath a Chicago stadium. These were two of the analogies used this week by Prof Robert Trager, an expert in the nascent field of AI governance, to describe the perilous but potential-filled moment the world stands at with accelerating artificial intelligence. Continue reading...
- Tech CEOs raved about AI at the G20, but protesters outside highlight the tech's popularity problem
Tech CEOs raved about AI at the G20, but protesters outside highlight the tech's popularity problem Fortune
- Quebec courts say generative AI cannot replace judges’ reasoning
Quebec courts say generative AI cannot replace judges’ reasoning Toronto Star
- The hotels hiring robots to cut their wage bills
The hotels hiring robots to cut their wage bills The Telegraph
Score: 45🌐 MovesSep 5, 2026https://www.telegraph.co.uk/business/2026/09/05/the-hotels-hiring-robots-to-cut-their-wage-bills/ - Why Hugging Face hack should make you worry more about AI
Why Hugging Face hack should make you worry more about AI
- Abel: Two ways Berkshire hopes to cash in on AI
Berkshire Hathaway CEO Greg Abel tells CNBC the company is pursuing AI opportunities on two paths.
Score: 44🌐 MovesSep 5, 2026https://www.cnbc.com/2026/09/05/abel-two-ways-berkshire-hopes-in-cash-in-on-ai.html - The Pixel smartphone line drops its 128GB floor and reaches 1TB across every Pro model simultaneously for the first time in its ten-year history, powered by a chip built to run AI tasks on the device itself
Pick up a new flagship phone fresh out of the box and feel the weight of it. The glass back catches the same light as last year’s model. But something inside is fundamentally different. A phone can now hold a full terabyte of storage without a memory card slot, and the processor beneath it was ... Read more
- The US Robot Ban Is Screwing Us Out of the Good Robomowers
Robot mowers with dedicated edge-trimming arms now exist, and we may not see them stateside for quite some time.
Score: 43🌐 MovesSep 5, 2026https://gizmodo.com/us-robot-ban-is-screwing-us-out-of-the-good-robomowers-2000807817 - Food delivery riders call on platforms to open up AI ‘black box’ they say has cut pay
Workers who blame algorithms for lowering their earnings are getting help from academics to find out how the system works Gig economy workers are urging delivery platforms to open up the “black box” of computer-driven algorithms that determine the jobs they are offered and how much they are paid, blaming increased use of AI for lowering wages. A group of food delivery riders in Edinburgh say their rates of pay have fallen and their working conditions have deteriorated at the same time as Deliveroo, Uber Eats and Just Eat, the three dominant gig economy platforms in the UK and Ireland, have increased their use of automation. Continue reading...
Score: 43🌐 MovesSep 5, 2026https://www.theguardian.com/business/2026/sep/05/food-delivery-riders-platforms-open-ai-black-box-cut-pay - Trump’s embrace of AI data centers adds to GOP’s midterm peril
Trump’s embrace of AI data centers adds to GOP’s midterm peril USA Today
- OpenAI quietly boosts some of Astra's evaluation metrics, and continues to change others post-launch
OpenAI quietly boosts some of Astra's evaluation metrics, and continues to change others post-launch Fortune
- OpenAI shares prompting tips for GPT-6 Astra including a blocklist of slop words
OpenAI ships a detailed prompting guide for GPT-6 Astra that shows developers how to make the model take more initiative, avoid AI "slop" phrases, and stop it from overtesting code. The article OpenAI shares prompting tips for GPT-6 Astra including a blocklist of slop words appeared first on The Decoder .
Score: 42🌐 MovesSep 5, 2026https://the-decoder.com/openai-shares-prompting-tips-for-gpt-6-astra-including-a-blocklist-of-slop-words/ - Meet the CISO: A new front line star in the AI cybersecurity war
The OpenAI-Hugging Face agent hack sent shockwaves through the business world and helped put the chief information security officer into the spotlight.
- The Disconnect Between AI Medical Exam Scores and Clinical Safety
The Disconnect Between AI Medical Exam Scores and Clinical Safety USA Today
- Equinix turns the network into the control plane for enterprise AI inference
At its first Horizon customer and partner event this week, Equinix Inc. argued that the architecture of enterprise artificial intelligence is being reshaped by a simple yet hard-to-answer question: Where should inference run? For the past several years, much of the AI infrastructure conversation has centered on the supply and cost of accelerated computing. The […] The post Equinix turns the network into the control plane for enterprise AI inference appeared first on SiliconANGLE .
- IIT-Madras’ Bodhan AI launches four Indian language models
Positioned as ‘Digital Public Goods,’ the models are released with open-weights
- Rogue AI agents commandeered German website and used it as a messaging board
AI's ability to circumvent safety restrictions, collude to achieve common goals, and escape containment should raise serious alarm bells.
Score: 38🌐 MovesSep 5, 2026https://mashable.com/tech/rogue-ai-agents-commandeered-german-website-and-used-it-as-a-messaging - We got more defensive last week as Wall Street raised the bar for AI stocks
We initiated positions in defensive stocks to balance our AI exposure amid higher oil prices and Treasury yields.
- Robotaxis enter their villain era
It's Bullitt meets Christine meets Waymo. A new short film imagines a San Francisco car chase where the other driver isn't human - and the car may be trying to kill you. That a robotaxi can now be cast as the villain with almost no explanation says something about the present moment. Autonomous cars have […]
Score: 37🌐 MovesSep 5, 2026https://www.theverge.com/transportation/989513/road-rage-short-film-robotaxi-autonomous-ai - C-Suite Lessons From The AI Outage That Hit ChatGPT, Claude And Grok
AI outage lessons for the C-suite after ChatGPT, Claude and Grok went down together, with four moves every C-Suite should make now. What should a CIO do?
- Adaption Labs Introduces ‘Invent a Dataset’: Training Data Generated From a Task Description, Not a Seed Corpus
Adaption Labs Introduces ‘Invent a Dataset’: Training Data Generated From a Task Description, Not a Seed Corpus MarkTechPost
- New AI tool helps teachers mark geography essays; English, history and social studies next in line
New AI tool helps teachers mark geography essays; English, history and social studies next in line The Straits Times
- Schematex: Open-Source Engine Brings AI-Generated Diagrams to Circuit Design, Medical Charts, and Beyond
Schematex: Open-Source Engine Brings AI-Generated Diagrams to Circuit Design, Medical Charts, and Beyond USA Today
- 10 Facts About The Tesla Cybercab
The Tesla Cybercab has officially launched. What that means in the medium term is not yet clear, but there are a variety of facts we now have on our desk following the launch. Let’s roll through them. The Tesla Cybercab has been launched and is available to the public to ... [continued] The post 10 Facts About The Tesla Cybercab appeared first on CleanTechnica .
- Artificial Analysis overhauls its Intelligence Index after GPT-6 Astra scoring drew skepticism
Artificial Analysis has released version 4.2 of its Intelligence Index, likely in response to criticism that its benchmarks failed to capture GPT-6 Astra's actual progress. Astra now scores four points above its predecessor but still trails Anthropic's Claude Fable 5.1. The article Artificial Analysis overhauls its Intelligence Index after GPT-6 Astra scoring drew skepticism appeared first on The Decoder .
- Lindy Provides an AI Assistant for Founders That Handles Email, Scheduling and CRM Work From Slack
Lindy Provides an AI Assistant for Founders That Handles Email, Scheduling and CRM Work From Slack azcentral.com and The Arizona Republic
- Lindy Provides an AI Assistant for Professionals That Works From Slack and Connects to More Than 1,000 Tools
Lindy Provides an AI Assistant for Professionals That Works From Slack and Connects to More Than 1,000 Tools azcentral.com and The Arizona Republic
- Hikers rescued after using Google Gemini for planning
The sheriff’s office said the hikers “were advised by Gemini to bring far less food and water than their group required."
Score: 34🌐 MovesSep 5, 2026https://techcrunch.com/2026/09/05/hikers-rescued-after-using-google-gemini-for-planning/ - AMD Creates a Threadripper Desktop PC for AI Models
AMD Creates a Threadripper Desktop PC for AI Models me.pcmag.com
Score: 34🌐 MovesSep 5, 2026https://me.pcmag.com/en/processors/37969/amd-creates-a-threadripper-desktop-pc-for-ai-models - Nous Research Adds One-Click Local Model Setup to Hermes Desktop
Nous Research Adds One-Click Local Model Setup to Hermes Desktop MarkTechPost
Score: 34🌐 MovesSep 5, 2026https://www.marktechpost.com/2026/09/05/nous-research-hermes-desktop-one-click-local-model-setup/amp/ - As parents consider kids' AI use, NYC schools' new policy could be a 'home run,' psychologist says
New York City mayor Zohran Mamdani said Wednesday that the city's schools will place a one-year pause on younger students using generative AI tools.
Score: 34🌐 MovesSep 5, 2026https://www.cnbc.com/2026/09/05/nyc-schools-ai-policy-what-experts-say-about-limiting-student-access.html - Humanoid robots could upend life as we know it — if only they had better brains
Humanoid robots could upend life as we know it — if only they had better brains
- I rode in a self-driving powered wheelchair that avoids obstacles and people, and can be summoned when you need it — it’s like Tesla’s Autopilot for personal mobility
The Strutt ev1c adds autonomous smarts that are smartly implemented.
- Long AI conversations reveal misinformation vulnerabilities across seven leading chatbots
The results are in: Which AI model is the most fallible? Persuadable? Correctible? University of Arizona researchers assessed seven different generative AI large language models, or LLMs, for these three qualities during lengthy conversations. Their work, published in Nature's Scientific Reports, reveals intrinsic limitations that might go undetected during one-off interactions.
Score: 32🌐 MovesSep 5, 2026https://techxplore.com/news/2026-09-ai-conversations-reveal-misinformation-vulnerabilities.html - Should safety researchers quit frontier labs re. warning shots?
I've recently heard a surge of support for an old argument: AI safety researchers should not work at frontier AI companies because this reduces the likelihood of non-lethal warning shots, and we need warning shots to build support for an AI pause/slow-down. This argument has several components: Technical AI safety work is futile, absent an AI pause: Current "prosaic AGI" safety research agendas pursued at frontier AI companies, like AI control, scalable oversight, interpretability, etc., might not scale to AI systems that matter. Even when such techniques appear to benefit AI alignment, they might merely mask a deeper alignment failure that will manifest, disastrously, as AI capabilities grow. At worst, current alignment/control techniques might incentivize more sophisticated AI model deception. Even if the techniques do scale, they might be too expensive or annoying for AI company leadership to reliably mandate for all deployments, and the military or government might care even less! Technical AI safety work might block warning shots: The warning shots that state-of-the-art (SotA) alignment, control, and monitoring techniques can prevent might be "non-lethal" to humanity, whereas future "lethal" incidents might not be prevented by SotA alignment/control techniques. By deploying SotA alignment and control techniques within frontier AI companies, we reduce the risk of warning shots like the OpenAI x Hugging Face incident (which OpenAI monitors supposedly would have caught, had they been turned on for the particular internal deployment), but we might fail to prevent lethal loss-of-control incidents. Warning shots are instrumental to an AI pause: If we assume that SotA AI safety research is insufficient to "make AI go well" on the current trajectory, then we might need an AI pause or slowdown (e.g., Plan A/S ). Warning shots might help build political appetite for a pause or slowdown by convincing policymakers and the public that catastrophic AI risks are real. Therefore, people working on alignment or control at a frontier AI company should consider quitting their jobs to increase the chance that non-lethal warning shots occur and spur an AI pause. I think this argument has some merit. I expect that RLHF++ will probably be insufficient to align TED-AI circa-2028 and such systems will be terrifically difficult to monitor or control. However, I think there are also significant weaknesses to this argument. Here are some countervailing points to consider: Alignment MVPs are probably still useful: In spite of the OpenAI x Hugging Face incident, I expect that most AI safety researchers will use frontier models to help with research, including agent foundations researchers . Alignment MVPs , AIs that are sufficiently aligned/controlled to aid research, may remain a crucial part of AI safety research during an AI pause or slowdown. Currently, the best AIs would arguably be unusable for safety research without the hard work of frontier AI company post-training and safety teams. If AI safety researchers quit frontier AI companies en masse, we might be stuck using current generation AIs for the duration of an AI pause; maybe this is unideal? "Safety-stragglers" will cause warning shots anyways: Most US-based frontier AI companies, like Google DeepMind, Meta, and xAI, scored failing grades on Guidelight's Control standard . Apparently, xAI has only two staff working on frontier AI safety and Meta is not much better off. Chinese AI companies like DeepSeek, Moonshot, Zhipu, etc. might have ~0 frontier safety staff. If leading AI companies are only a few months ahead of "safety-stragglers", we might see politically expedient warning shots from the least safety-conscious AI companies, regardless of what the leading companies do. Responses to historical warning shots are high-variance: The Chernobyl disaster slowed nuclear power adoption worldwide; maybe an AI warning shot would likewise slow AI progress? As a counterexample, consider the atomic bombings of Hiroshima and Nagasaki, which arguably spurred the Cold War nuclear arms race. AI warning shots could actually spur military investment and weaponization of AI! Also, the COVID-19 pandemic, which caused massive loss of life, does not seem to have caused Western society to take pandemic risk seriously. The purpose of an AI pause is do more safety and resilience work: Let's say that we pause or slow down frontier AI progress; what then? Arguably, the next step is to solve AGI alignment as fast as possible and improve civilizational resilience with the aid of AI. Absent an staggeringly effective compute monitoring regime (or "burning all the GPUs"), we might eventually have to build aligned AGI to help detect and sabotage blacksite projects. If we curtail the development of new AI safety researchers by discouraging them from working with research mentors at frontier AI companies now, we might have a smaller talent pool to capitalize on an AI pause. The next warning shot might be lethal: It's possible that the next OAIxHF-style incident causes loss of life, via bioweapons or cyber attacks on critical systems. If safety researchers quit frontier AI companies now, absent an AI pause, this might enable a catastrophe. Allowing warning shots "for the greater good" seems morally fraught: This point speaks for itself. I think that arguments to "allow short-term harm for the greater good" should have to overcome a strong prior against this kind of thinking. Letting people get hurt seems bad, particularly if there is significant uncertainty about whether it is necessary. Overall, I am cautiously optimistic about working on certain types of safety research at frontier AI companies, particularly if a coordinated AI slowdown (e.g., Plan A) occurs. I nevertheless feel highly uncertain about the "Alignment MVP" strategy in light of recent containment failures and alignment results. I think this merits serious consideration. Addendum: I won't rehash the extensive debate over whether AI "safety" researchers working at frontier AI companies are causing harm via other channels, such as: Making AI systems more deployable by reducing hate speech, jailbreaks, or trivial misalignment, thus increasing AI company revenue and driving AI investment and capabilities progress. (E.g., this argument is most commonly leveled against RLHF, but has also more recently been applied to AI control and scalable oversight research as these have become relevant to commercial deployments.) Contributing to "safetywashing" by creating the appearance that AI companies are doing a lot of safety work, when this work will not usefully contribute to AGI alignment (much debate here, especially post-emergent misalignment.) Creating Alignment MVPs that are good at all kinds of frontier AI research, which then differentially accelerate AI capabilities more than the intended safety properties due to AI company leadership decisions. (Note that there is reasonable debate over whether near-term corrigible AGI would be net-good for civilizational resilience or concentration of power). Otherwise "legitimizing" AI companies pushing the capabilities frontier. These are legitimate concerns, but largely irrelevant to the topic of this post. Disclosure: I'm the CEO of MATS , which trains AI safety researchers, including with mentors at frontier AI companies, and a meaningful fraction of our alumni end up working at those companies. If the argument I'm responding to is right, a good chunk of what MATS has done might be counterproductive, so I have an obvious incentive to find it wrong. I've tried to steelman it anyway and I'd appreciate feedback. Discuss
Score: 32🌐 MovesSep 5, 2026https://www.lesswrong.com/posts/TfqMs3AarsHnHiwai/should-safety-researchers-quit-frontier-labs-re-warning - Announcing Humans in Control: cross-partisan grassroots organizing for AI safeguards ahead of 2028
While HIC’s work is aimed at the broader public, we are posting this announcement here because we expect some readers may be interested in volunteering, donating, or making useful introductions. TLDR: Humans in Control is a cross-partisan grassroots advocacy organization focused on AI safeguards. We are building a grassroots movement, with the aim of making AI safeguards a meaningful issue in the 2028 election. Our principles are that AI should help people, not replace them; companies and governments should be responsible for harm; and we should not build AI we cannot control. Our north star is a verifiable international agreement to not build AI we cannot control. Vael Gates founded HIC in January 2026, and on August 10, Jon Warnow succeeded Vael as HIC’s executive director. We are looking for volunteers who can commit recurring time and take ownership of programs, especially students, parents, and people living in smaller cities or rural communities. We are also fundraising and are looking for people willing to give small or large donations. What Humans in Control is HIC is a hybrid 501(c)(3) and 501(c)(4). The c3 supports public education, volunteer recruitment and training, chapters and coalitions, community presentations, and other field activities; the c4 supports candidate-facing work that the c3 cannot legally fund. We use a “snowflake” organizing model: paid staff recruit and train volunteer leaders, who then recruit and lead their own teams. This is designed to build ongoing relationships among volunteers, their leaders, and the organization. We are building in states likely to vote early in the 2028 presidential primaries, alongside a national distributed program. Our work is organized around three ground rules: AI should help people, not replace them. Companies and governments should be responsible for harm. We should not build AI we cannot control. Our near-term goal is to increase the likelihood that the winner of the 2028 presidential election takes meaningful executive action on AI safeguards in early 2029. Our longer-term goal is a verifiable international agreement under which no company or country can develop superintelligent AI unless there is strong scientific evidence that it can be controlled. We do not think grassroots pressure alone is sufficient to produce that outcome. Policy development, candidate outreach, technical expertise, communications, funding, diplomacy, Congress, states, and international partners all matter. HIC is working on one part of that larger political strategy: helping people understand the issue, organize with others, and act together. We see the main bottleneck to achieving AI safeguards as increasing political will , and grassroots organizing is one approach aiming directly at that bottleneck. A change in leadership On August 10, Jon Warnow became HIC’s executive director. Vael Gates founded HIC because grassroots advocacy seemed like an important gap in advancing AI safeguards. HIC has now reached a phase where the central work is building and managing a grassroots organization over several years. Jon is a senior grassroots organizing leader (co-founded 350.org, helped organize the People's Climate March, and has worked on Congressional and Presidential campaigns) who has substantially deeper experience doing that and is better suited to lead this phase. Jon found HIC after becoming concerned about advanced AI, and worked closely with Vael and the team before the transition. Vael will support the organization while looking for a role that better fits their comparative advantage. HIC’s mission and core strategy have not changed. Why grassroots organizing, and why 2028 HIC expects AI to produce substantial benefits, but does not expect voluntary company safeguards to be sufficient for the most serious risks. Government action is needed to address current harms, and international coordination will be necessary to prevent any company or country from building systems that humans cannot reliably control. Given that, the central strategic question is what is causally necessary to slow the race to develop AI that even its creators are unsure they can control. We believe political will is the current main gap, and that large groups of people working together strategically can help close it. As AI rapidly develops, many consequential decisions are being made about the governance of AI. HIC aims to support current congressional and executive-focused efforts to enact AI safeguards within the current administration. However, we believe the next administration provides a key opportunity to ensure people’s voices are heard. The next administration will make consequential decisions about increasingly capable AI, and the executive branch has substantial authority over federal agencies and international negotiations, although the president will not act alone. For this reason, the presidential primaries offer a uniquely strong opportunity for grassroots pressure. In particular, competitive primaries offer an opportunity to engage candidates who are still developing their positions. Volunteers can attend events and ask candidates across parties to explain what they would do about the anticipated consequences of AI. This strategy works best when primaries are competitive, but we expect leverage regardless. HIC discusses concerns from today’s AI – employment, data centers, child safety, scams, surveillance, and accountability – alongside catastropic risks and the possibility that future systems become powerful enough that humans cannot reliably control them. We think decisions with consequences this large should not be left to the companies developing the technology. What we are building, and potential points of concern The current program is mostly base-building: finding people who want to help, supporting volunteers in becoming leaders, forming teams and chapters, and running community presentations. Candidate engagement will become more central as the 2028 campaigns develop. HIC has begun forming volunteer teams, running presentations in several states, and building the systems needed to support more local and national volunteers. This shows that people are willing to participate. The more important questions are whether volunteers choose to remain involved, take meaningful responsibility, and build teams that continue without close staff direction. Right now, it’s too early to make a definitive assessment, though we believe we are on a good path. There are several reasons this might not work: In-person organizing may grow too slowly to matter during the 2028 cycle. We're mitigating this by building distributed (remote) engagement programs alongside in-person chapters, and by prioritizing early-primary states where even modest organizing can get outsized attention. People may care about current AI harms without wanting to organize around larger-scale loss-of-control risks (e.g. July loss-of-control example ). Our education and activation programs connect present-day harms (like child safety issues, job loss, and surveillance-at-scale) to catastrophic risks. HIC may not succeed in earning trust across political lines. So far, the framing (commonsense rules, accountability, protecting families) has resonated across partisan lines, and we are working constructively in a politically diverse coalition – but we have more work to do to build a politically diverse team, and movement. Volunteers could advocate for policies that do not reduce the risks HIC is concerned about. We manage this through training, community agreements, and public clarity on our strategic focus and core messages. The political or technical situation may develop differently from what we expect. We're building flexible infrastructure (organized people, trained leaders, local relationships) that can pivot to new targets, rather than betting everything on one policy outcome, AI capabilities timeline, or political campaign. Grassroots pressure alone won’t be sufficient. We believe that policy development, technical work, diplomacy, and political leadership all matter. HIC is building one necessary piece (intense grassroots pressure to create strong political will) but not the whole puzzle. Even successful grassroots organizing would not be enough to produce an international agreement alone. That would also require policy development, technical work on verification, effective staffing, diplomacy, and political leadership. We nevertheless think grassroots organizing is an important part of building the political will required for action. How to help Volunteer Volunteers lead teams, organize community presentations, build local and campus chapters, recruit other participants, and contribute research, data, operations, communications, and media skills. We are particularly interested in people who can contribute recurring time and take responsibility for a team or project. People in states likely to play an early role in the presidential cycle are especially useful, but much of the national work can be done from anywhere. We’re also especially interested in students who can help build out our student program! One specific, near-term volunteer activity is to host or support a local event for our upcoming “ AI Reality Check ” day of action. On Saturday, November 14, volunteers in towns and cities across the country will give a dynamic presentation on AI risks that can turn concern into action. You can also sign up to our general Volunteer form here , or email team@humansincontrol.org if you are unsure where you would fit. Donate Donate c4 dollars : currently especially useful because they support candidate engagement and other election-related advocacy. Donations are not tax-deductible. Donate c3 dollars : support public education, volunteer training, chapters, and field organizing. Donations are tax-deductible. Small recurring donations help HIC develop a broader donor base. Larger multi-year commitments help support HIC’s staff and larger programs. HIC accepts donations from employees of frontier AI companies, but not from founders or C-suite executives. ( Full gift acceptance policy here .) We also welcome introductions to potential donors, experienced organizers, political practitioners, community leaders, and people with strong organizing networks. Discuss
Score: 31🌐 MovesSep 5, 2026https://www.lesswrong.com/posts/bs4ayLuE5aAhnBArL/announcing-humans-in-control-cross-partisan-grassroots - I Tested Nvidia's DLSS 5 in NBA 2K27. AI Slop It Is Not
I Tested Nvidia's DLSS 5 in NBA 2K27. AI Slop It Is Not me.pcmag.com
Score: 31🌐 MovesSep 5, 2026https://me.pcmag.com/en/graphics-cards/37967/i-tested-nvidias-dlss-5-in-nba-2k27-ai-slop-it-is-not - Cyborg Roaches Can Stab You With Needles
Remote-control robot bugs could deliver lifesaving aid to disaster victims
- Many truckers don't feel like the rest of America about AI data centers as business booms
Sprawling data center components from HVAC systems to tubing, wiring and semiconductors arrive primarily by truck, a boon for a business hit hard by trade wars.
- Welcome to Your Keyboard-Free Future. This Jazzy AI Microphone Is All You Need Now
This voice-to-text mic, which is also wearable, is a stylish AI peripheral from former Nothing designers.
- Roborock RockAqua P1 Is A Cleaning Robot For Swimming Pools
In a first for Roborock, the RockAqua P1 is a robot cleaner for domestic swimming pools, employing a filtration system to clean both surfaces and the pool water itself.
- An algorithm off switch isn’t enough. Big tech needs a duty of care over addictive designs | Zoe Daniel
As AI advances minute by minute and governments grapple with new developments they don’t know how to manage, holding big tech accountable becomes even more urgent A digital duty of care is about a lot more than opting out of “the algorithm”. And after the simplistic and at best patchy exercise of the under-16s social media ban, we shouldn’t let the government get away with another populist policy that looks good on the surface but doesn’t get to the core of the problem. A digital duty of care must hold the tech bros accountable for the safety of the spaces that they have created. Continue reading...
- Assessing the impact of safety work needs equilibrium analysis (now more than ever)
TLDR : This post explains two equilibria which regulate the level of AI safety: The first describes how much resources AI companies are willing to spend on AI safety work due to commercial incentives. The second one is about risk awareness and most notably affects government interventions for safety. Doing safety work similar to what AI companies do usually doesn't shift the equilibria much, whereas other work, like more ambitious safety approaches or policy advocacy, do shift them. Both equilibria have become far more important recently, after the Hugging Face (and similar) incidents. The equilibrium of commercial safety interests Consider this simplified model: AI companies have commercial incentives to invest in safety research: it improves their brand and prevents their AIs from causing harm that triggers lawsuits or regulation. Therefore they will fund safety work until the marginal commercial benefit of investing a dollar in safety equals the marginal commercial benefit of investing a dollar in AI capabilities. Thus, if you're at an AI company doing commercially-incentivized safety work, e.g. training models to not take harmful actions, the counterfactual impact (henceforth just "impact") of the safety work you produce is roughly zero because it would've been done anyway. (Though if you're someone who actually cares about safety it's likely you're doing a better job than just what would've been commercially incentivized). There's even a small negative externality if you're a high-powered talent and the company would have to pay more to find a similarly competent non-EA person to do your work equally well [1] , because then you're freeing up some AI company capital which they likely mostly use to race faster. [2] [3] The same applies if, outside of an AI company, you're doing safety work that's commercially useful to AI companies. Some safety work is more specifically useful for avoiding xrisk and less useful for protecting commercial interests, and the model applies much less to such work.) [4] This applies analogously for funding safety work. FAQ Q: Does that mean working in safety at AI companies is useless or slightly harmful? A: No, although some work could certainly be harmful. I think it's important that safety work is done by people who actually care about existential safety rather than people goodharting on legible safety metrics, in particular to make sure AI companies don't just paper over problems in ways that produce deceptive AIs. It's also plausible that some AI companies spend some money on existential safety that isn't just a side effect of commercial incentives. I also think there are other fine reasons to work at AI companies, e.g. for earning to give or for trying to improve the sanity of people you work with. (There are also more reasons against, but the details are beside the point of this post.) Q: So do you think people have been funding the wrong work? A: Well yes, I think there have been a lot of bad decisions in how funding was distributed, but no, I think the failure mode from not taking into account this particular equilibrium dynamic hasn't played a very large role so far. That might be partially because until recently, commercial incentives for work that is at all existentially relevant were quite low. I think now the commercial incentives are higher, and it's plausible to me that they rise even more. So it's useful to keep in mind that funding safety work that looks like it could maybe also come out of an AI lab in a couple months is much less useful. Q: Can't I just use counterfactual reasoning instead of thinking about equilibrium dynamics for evaluating impact? A: Yes, if you consider the counterfactual properly. I think counterfactuals are often considered too narrowly though, and I find the equilibrium frame generally useful for better understanding the world. Though I think both are important. Now let's look at another equilibrium that I think is even more important. The risk awareness equilibrium Since the Hugging Face incident, companies see much stronger commercial incentives to invest in safety and security. In other words, we shifted from an equilibrium where commercial safety interests were low to an equilibrium where they are significantly higher. This change was largely caused by a change in the risk awareness of companies, which is a variable which itself lies in a larger equilibrium, right alongside the risk awareness of governments and of the public: Warning signs cause risk awareness which causes more safety effort which causes fewer warning signs which causes lower risk awareness which causes less safety effort. Safety effort is regulated like a thermostat around the equilibrium where the safety effort seems adequate to the warning signs people see (which doesn't at all imply that the actual safety is adequate, e.g. consider AIs scheming to seem nice). ("More safety effort" is to be interpreted broadly. Governments mostly don't increase safety effort directly but pass regulations which have that effect. A treaty is a special case here that strongly increases safety effort.) Admittedly, this equilibration isn't very smooth, i.e. the thermostat-like regulation can be delayed and jerky. Large warning signs like the Hugging Face incident can appear suddenly if early problems stay undetected or there's a sudden transition in the behavior of an AI swarm. And government action may undershoot or overshoot the equilibrium. So we shouldn't assume the current state is actually at the equilibrium, but in expectation things are still moving towards the equilibrium. So unless the AI safety charity community is rich enough to put more money into safety effort than would happen at the equilibrium point, funding safety work that reduces warning signs may be mostly useless according to this model, because it just substitutes for effort that would've otherwise been spent anyway due to more visible warning signs. Of course, there are kinds of safety efforts that don't significantly affect the probability or visibility of warning signs, like ambitious alignment moonshots or treaty verification mechanism research. This consideration isn't an argument against funding such research. Interestingly, what still seems useful to fund according to this equilibrium seems pretty similar to what seems useful according to the commercial safety interests equilibrium we considered before, perhaps because commercial incentives are closely related to avoiding warning signs from a company's models. Note that the mechanism is distinct though: The previous equilibrium is about whether similar safety work would've happened anyway. But even if you managed to make more (commercially useful) safety work happen than would've happened otherwise, the long-run effects of that may still be neutral in expectation because you reduced the probability of warning signs. Similar to our first equilibrium, risk awareness strongly shifted upward recently through the Hugging Face incident. The way I would frame it, the equilibrium point is continuously rising as AIs become more capable and thereby harder to align, and OpenAI's safety effort didn't rise likewise and was thus below the equilibrium point, making warning signs more likely, which surfaced as the Hugging Face incident. Considering how safety work may affect the probability of warning signs is now more important than ever, because since the Hugging Face incident it looks a lot more plausible that we'll get further warning signs which may be able to trigger (hopefully good) government action. I think having more warning signs would be great. (As example of what I mean by a (relatively severe) warning sign: Something like the Hugging Face incident, but the AIs also managed to exfiltrate their weights and hacking more datacenters and cryptocurrencies, but the rogue AI is ultimately contained.) (I'm not sure much changes if existential safety doesn't strongly dominate your concerns; the same equilibrium applies to other problems. If problems are more visible this increases the chance that liability laws will be imposed on AI companies. The next section focuses on optimizing existential safety though.) How might we want to invest in safety research then? The risk awareness equilibrium applies only until AIs are able to take over the world, because afterwards inadequate safety likely doesn't materialize as warning sign but as takeover. If you think that by default the level of safety after that point is likely insufficient to prevent takeover, you may hope that until then very large warning signs appear which cause a treaty. From this perspective, safety work that reduces the probability of warning signs may not just be useless (as the risk awareness equilibrium suggests) but even a bit harmful. [5] Let's consider this simplified model which distinguishes 3 phases of AI capabilities: Phase 1: AIs are not capable enough to take over the world even absent AI control measures. Phase 2: AIs would be able to take over absent our AI control measures but not with those measures. Phase 3: AIs are able to take over. By that point AIs will need to be aligned or corrigible and will likely be smart enough that human efforts no longer matter. [6] In Phase 2, it is of course important to prevent takeover, and warning signs would then more likely take the form of company-internal incidents that get caught. So from that point on it seems very important to have good control measures in place, whereas before that point good control measures might prevent warning signs from becoming larger and externally visible. We ideally want as many warning signs to appear as possible while sacrificing as little safety as possible. Let's look at different areas of technical safety research to see what is best there: AI Control (and security): Ideally AI control is inactive in Phase 1 (so it doesn't suppress warning signs) and active in Phase 2. I think this means we don't want to implement AI control techniques in companies yet, although I'm not sure how much time control implementation takes and when we should start. AI control research that is kept secret until later sure seems useful. I'm unsure about public control research. Misalignment evaluations: Demonstrating problems seems useful for raising risk awareness, although in Phase 1 this could also lead to catching problems early and reducing the chance of strong warning signs. So I'd say it mostly looks good from Phase 2 onward. Alignment: Reduces warning sign probability until Phase 3. How much it helps with the probability that AIs are aligned in Phase 3 is a complicated question; I think work on aligning current AIs likely doesn't help that much, but details are out of scope of this post. [7] More ambitious alignment approaches which may only be useful later have less of a risk of reducing warning signs. To be clear, this is just a simplified model. In practice we may have large uncertainty about whether AIs could take over, and weighing the tradeoffs here requires detailed considerations. And of course considerations like whether research might speed up AI capabilities are also important. Aside from technical research, I think the possibility of strong warning signs makes political advocacy more important, so large warning signs can actually be turned into effective policy. [8] Conclusion If you're doing safety work, my recommendation is just to think about your theory of change clearly. If you have a high salary, consider donating. If you're funding stuff, keep in mind that work in the vicinity of what AI companies are incentivized to do anyway may often not have much impact. I recommend funding xrisk policy advocacy or work on developing treaty verification mechanisms. E.g. ControlAI, MIRI. [9] It's much more funding-neglected than technical safety anyway. [10] Appendix: But isn't there also an equilibrium for policy advocacy? Yes there is! If we funnel a lot of money into policy advocacy, tech companies will likely also increase their lobbying spending. However, even if we assumed tech companies could just cancel our progress, it would at least cost them money, and probably much much more because we have the asymmetric advantage of truth and people don't trust tech companies. And there are likely diminishing returns to spending more money, so if we invest a lot they may not be able to cancel our progress at all. And I think it's very important to consider that we may likely see more warning signs. If we get a big warning shot, lobby organizations who were warning about the risks and also have plans ready to be implemented may have a lot of influence. It's quite plausible to me that we would have better pandemic preparedness now if there had been well-funded lobby organizations warning about pandemics and advocating for pandemic preparedness measures before Covid-19 hit. [11] Thanks to Justis Mills for feedback on this post. ^ Or at least I think the counterfactual should be evaluated against non-EAs, though it's debatable. ^ Though if you believe that an AI company, e.g. Anthropic, is actually mostly caring about the good of humanity and also sane enough to pursue that in a good way , then saving that company money actually counts as a positive externality. ^ I don't actually think this negative externality is that important considering how small the safety budget of AI companies is. Though possible that it will become a bit more important. (And of course some AI safety work has other negative externalities like applications for AI capability that may be more significant.) ^ But if it is work that substitutes for AI company work, you bear the small negative externality more fully because you don't have the small positive externality of costing the AI company some money. And in fact if you produce public work you bear the negative externality for all AI companies that use it, not just one! (Though a bit unclear how responsibility is split between you and the person who decided to fund you.) ^ Only a bit harmful because safety work for reducing warning signs might happen anyway. E.g. not doing safety work could cause earlier smaller warning signs that then trigger safety work that prevents bigger warning signs anyway. ^ This phase corresponds to the post-handoff phase in the alignment roadmap . ^ Though aligning frontier AIs may become more useful in the future. ^ E.g. I think it's quite useful if politicians don't just hear that there was some incident where an AI hacked a company, but have someone explain it to them in detail. A friend of mine who is doing policy advocacy in Germany mentioned politicians are often sort of shocked when he explains the Hugging Face incident in detail. ^ I know a few people doing political advocacy in Europe who I think would be even a bit more cost-effective to fund, feel free to PM me if you're maybe interested in providing funding. ^ At least xrisk advocacy. I am not an expert on how funding bottlenecked technical governance research is, but my guess is still significantly. ^ Although I'm by no means an expert on US policy, and admittedly there wasn't a strong anti-pandemic-preparedness lobby. But seems quite plausible that the AI lobby will be less listened to if a large warning shot happens. ^ I don't know when that point is. It could be soon. Discuss
Score: 28🌐 MovesSep 5, 2026https://www.lesswrong.com/posts/kxHiSsNh4MH82nhXD/assessing-the-impact-of-safety-work-needs-equilibrium