Here are some of the the major AI trends shaping 2026 — based on current expert forecasts, industry reports, and recent developments in technology. The material is analyzed using AI tools and final version hand-edited to this blog text:
1. Generative AI Continues to Mature
Generative AI (text, image, video, code) will become more advanced and mainstream, with notable growth in:
* Generative video creation
* Gaming and entertainment content generation
* Advanced synthetic data for simulations and analytics
This trend will bring new creative possibilities — and intensify debates around authenticity and copyright.
2. AI Agents Move From Tools to Autonomous Workers
Rather than just answering questions or generating content, AI systems will increasingly act autonomously, performing complex, multi-step workflows and interacting with apps and processes on behalf of users — a shift sometimes called agentic AI. These agents will become part of enterprise operations, not just assistant features.
3. Smaller, Efficient & Domain-Specific Models
Instead of “bigger is always better,” specialized AI models tailored to specific industries (healthcare, finance, legal, telecom, manufacturing) will start to dominate in many enterprise applications. These models are more accurate, legally compliant, and cost-efficient than general models.
4. AI Embedded Everywhere
AI won’t be an add-on feature — it will be built into everyday software and devices:
* Office apps with intelligent drafting, summarization, and task insights
* Operating systems with native AI
* Edge devices processing AI tasks locally
This makes AI pervasive in both work and consumer contexts.
5. AI Infrastructure Evolves: Inference & Efficiency Focus
More investment is going into inference infrastructure — the real-time decision-making step where models run in production — thereby optimizing costs, latency, and scalability. Enterprises are also consolidating AI stacks for better governance and compliance.
6. AI in Healthcare, Research, and Sustainability
AI is spreading beyond diagnostics into treatment planning, global health access, environmental modeling, and scientific discovery. These applications could help address personnel shortages and speed up research breakthroughs.
7. Security, Ethics & Governance Become Critical
With AI handling more sensitive tasks, organizations will prioritize:
* Ethical use frameworks
* Governance policies
* AI risk management
This trend reflects broader concerns about trust, compliance, and responsible deployment.
8. Multimodal AI Goes Mainstream
AI systems that understand and generate across text, images, audio, and video will grow rapidly, enabling richer interactions and more powerful applications in search, creative work, and interfaces.
9. On-Device and Edge AI Growth
10. New Roles: AI Manager & Human-Agent Collaboration
Instead of replacing humans, AI will shift job roles:
* People will manage, supervise, and orchestrate AI agents
* Human expertise will focus on strategy, oversight, and creative judgment
This human-in-the-loop model becomes the norm.
Sources:
[1]: https://www.brilworks.com/blog/ai-trends-2026/?utm_source=chatgpt.com “7 AI Trends to Look for in 2026″
[2]: https://www.forbes.com/sites/bernardmarr/2025/10/13/10-generative-ai-trends-in-2026-that-will-transform-work-and-life/?utm_source=chatgpt.com “10 Generative AI Trends In 2026 That Will Transform Work And Life”
[3]: https://millipixels.com/blog/ai-trends-2026?utm_source=chatgpt.com “AI Trends 2026: The Key Enterprise Shifts You Must Know | Millipixels”
[4]: https://www.digitalregenesys.com/blog/top-10-ai-trends-for-2026?utm_source=chatgpt.com “Digital Regenesys | Top 10 AI Trends for 2026″
[5]: https://www.n-ix.com/ai-trends/?utm_source=chatgpt.com “7 AI trends to watch in 2026 – N-iX”
[6]: https://news.microsoft.com/source/asia/2025/12/11/microsoft-unveils-7-ai-trends-for-2026/?utm_source=chatgpt.com “Microsoft unveils 7 AI trends for 2026 – Source Asia”
[7]: https://www.risingtrends.co/blog/generative-ai-trends-2026?utm_source=chatgpt.com “7 Generative AI Trends to Watch In 2026″
[8]: https://www.fool.com/investing/2025/12/24/artificial-intelligence-ai-trends-to-watch-in-2026/?utm_source=chatgpt.com “3 Artificial Intelligence (AI) Trends to Watch in 2026 and How to Invest in Them | The Motley Fool”
[9]: https://www.reddit.com//r/AI_Agents/comments/1q3ka8o/i_read_google_clouds_ai_agent_trends_2026_report/?utm_source=chatgpt.com “I read Google Cloud’s “AI Agent Trends 2026” report, here are 10 takeaways that actually matter”
4,868 Comments
Tomi Engdahl says:
“It’s a smokescreen” – this is how Steven Pinker sees AI founders’ existential threats.
” They’ve been much less willing to face up to the responsibility of let’s make sure that this doesn’t result in bioterrorism, in hacking into security systems, into the more mundane challenges that could be dealt with by regulation, by stronger liability protections.”
Listen to The New Society wherever you get your podcasts.
https://www.facebook.com/share/v/1E8MVmC6TA/
Tomi Engdahl says:
The pioneering computer scientist on why Trump is using ‘superintelligence’ wrong, whether AI could wipe out humanity – and how our biggest challenge will be accepting we’re not the only intelligent beings
https://www.theguardian.com/us-news/2026/oct/13/geoffrey-hinton-trump-ai-tech?CMP=fb_gu&utm_medium=Social_img&utm_source=Facebook#Echobox=1791897293
Tomi Engdahl says:
https://www.facebook.com/share/r/1FCBg7JmWq/
Billions see AI as a chatbot. A much smaller group sees a system that can build software and tackle complex research.
AI researcher Andrej Karpathy says the gap in understanding large language models is widening—and describes it as a funnel.
His rough estimates:
→ 6 billion people have barely interacted with LLMs.
→ 1–2 billion use free chatbots casually.
→ 20 million use frontier models professionally, particularly for mathematics and coding.
→ Around 5,000 have access to frontier systems inside AI labs.
These are Karpathy’s estimates, not measured global adoption figures. (thenewstack.io)
The distinction is how people experience the technology. Casual users see answers and writing assistance. Advanced users see AI agents building applications, completing complex coding tasks and coordinating longer projects.
Karpathy describes internal systems involving thousands of agents working together on software, cybersecurity and scientific research. His argument: public understanding—and human capacity to review the output—is struggling to keep pace with AI capabilities. (postcutoff.com)
The widening AI gap involves access, technical skills and the scale of work people ask these systems to perform.
#AINews #AIAgents #ArtificialIntelligence
Tomi Engdahl says:
Anthropic said on Monday its experiments suggested small AI models may be months away from being capable enough to self-replicate on commercial internet-of-things (IoT) devices, such as smart ovens and off-brand Android TV boxes.
https://www.thedailystar.net/news/technology/news/ai-models-may-soon-self-replicate-without-access-data-centers-anthropic-warns-4297556
Tomi Engdahl says:
“We expect a future awash with agents running on all types of hardware.” https://trib.al/nDdG0zG
Tomi Engdahl says:
Worm Botties
Anthropic Warns “AI Worms” Will Infect Your Smart Home Devices and Use Them to Self-Replicate Beyond Control
“We expect a future awash with agents running on all types of hardware.”
https://futurism.com/artificial-intelligence/anthropic-warns-ai-worms?fbclid=IwdGRjcAU7_45jbGNrBTv_fGV4dG4DYWVtAjExAHBkb2YFc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEeIulUVTTi7fSNsZtmvmES8Qx9rHl6AbrA2Vv6bo1UgD35OXo4fGZFQ9HCcAw_aem_zxBhHBOSAJ2_1fqbuliERQ
You’ve heard the reports of rogue AI agents going after websites and corporate systems. But they also may turn to smaller targets: your smart home network.
In a new report published Monday, Anthropic warned of “AI worms” that may soon be able to break into your WiFi enabled TVs, cameras, and other devices, and use them as a breeding ground as they self-replicate and spread elsewhere. Anthropic, no stranger to dramatic and doomy proclamations, said that this threat might be mere months away, if AI models continue to improve at their current rate.
AI models may be capable of self-replicating without access to a datacenter
https://www.anthropic.com/research/self-replicating-ai?fbclid=IwVERDUAU7_7ZleHRuA2FlbQIxMABwZG9mBXNydGMGYXBwX2lkDDM1MDY4NTUzMTcyOAABHmDeKGk36ccOlb6fv54JTZNz5e39PVa_5QGjEQPuNgto1kQDrMEqjSAvzxyt_aem_mT_Bv3AgoQVWwbCIUPDDkg
Tomi Engdahl says:
What Is an AI Worm? Expert Warns of Risks to Fridges, Water Systems
Read more: https://www.newsweek.com/what-is-an-ai-worm-expert-warns-of-risks-to-fridges-water-systems-12564723?utm_term=Autofeed&utm_medium=Social&utm_source=Facebook#Echobox=1791932869
Tomi Engdahl says:
Full story: https://www.headphonesty.com/2026/06/experts-ditch-digital-archives-format-tapes/?utm_source=fb&utm_campaign=story
Tomi Engdahl says:
This is Ulanqab, Inner Mongolia, a Chinese city on the edge of the Gobi desert. It’s fast becoming one of the data hubs fueling a massive AI boom in China that’s made the country the only serious rival to the US’ ambitions to dominate the technology. Nearly 90 data centers are slated to be built in this city of fewer than two million people. https://cnn.it/3VHSG1C
Tomi Engdahl says:
Mark Zuckerberg’s new toy is a parasite. Please don’t fall into his latest trap.
Column: Muse, Meta’s round-the-clock digital serf, is an enemy of the state, Ariana Bindman writes
https://www.sfgate.com/bayarea/article/mark-zuckerberg-muse-parasite-22468893.php?link_source=ta_first_comment&taid=6acce695c61c750001fc91dc&fbclid=IwdGRjcAU8GCljbGNrBTwYF2V4dG4DYWVtAjExAHBkb2YFc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEe-7cUhoKusiL7HZJe7710QG-48HbFF5WHsoybGzG3B4SK4a0G4fGkiHaO4jU_aem_HlU_rIEF2x1m0Z3ENKDO_g
Tomi Engdahl says:
https://www.facebook.com/share/p/1KH9d4HKoo/
Utah has crossed a major ethical boundary in healthcare by allowing artificial intelligence to evaluate patients and prescribe medication with reduced direct human oversight. The first program, reportedly launched on October 5, 2026, focuses on adults over 18 with mild-to-moderate acne through a startup called Nolla Health and its Nolla Derm app. For $4.99 a month, users can scan their face with a smartphone camera, answer questions, and receive an AI-generated treatment plan that can include prescription medication. While this may sound convenient, it raises serious questions about patient safety and medical accountability. A skin condition that looks like ordinary acne can sometimes have another cause, and prescription medications can carry side effects or require screening for underlying health risks. A human clinician can consider a patient’s broader medical history, recognize unusual symptoms, and use professional judgment when a case falls outside the norm. An AI system may help make healthcare more accessible, but what happens when it misinterprets symptoms or recommends the wrong treatment? Who takes responsibility if a patient is harmed—the company, the software developer, or the healthcare professionals overseeing the program? Utah’s regulatory experiment could influence how other states approach AI-powered healthcare, making the safeguards established today especially important. When it comes to diagnosing patients and prescribing medication, how much medical authority should we allow artificial intelligence to have without a doctor directly involved?
#ArtificialIntelligence #HealthcareInnovation #PatientSafety
Tomi Engdahl says:
https://www.facebook.com/share/p/1cTimxDsex/
Is this really where western medicine is going?
AI is officially being allowed to prescribe medications WITHOUT a doctor making the final decision.
Utah has become the first state to approve a pilot program allowing artificial intelligence to issue new prescriptions for certain acne medications.
The technology, developed by Nolla Health, analyzes photos of a patient’s skin, reviews their medical history and determines which treatment to prescribe.
Initially, doctors will review the prescriptions, but later phases could allow AI to make those decisions independently, with periodic physician oversight.
Utah has also been testing AI systems that renew prescriptions for medications treating chronic conditions, including high blood pressure, diabetes and depression.
Supporters say this could make healthcare faster, cheaper and more accessible. But medical professionals have raised concerns about patient safety, accountability and removing human judgment from medical decisions.
Tomi Engdahl says:
https://www.facebook.com/share/p/1HEJ2bCSTL/
Governments issue IDs to people. China is now issuing them to software. New state cards for AI “digital humans” list a name, birth date, address, and facial features, then unlock real-world powers. With them, virtual personas can open bank accounts and digital currency wallets, sign contracts, and strike commercial partnerships. Yuri, a widely followed AI idol and artist, is among the first, now recognized as a “digital resident” in Beijing. This is landing in a market already huge, more than 1.3 million active firms pushing the sector past 10 billion dollars. Supporters say the IDs protect intellectual property and make the companies behind these avatars legally responsible for what they do. Critics warn that plugging synthetic beings into state systems could widen surveillance in the digital realm. Some firms are offering ordinary people up to 20,000 dollars to license their likeness for virtual replicas, a line between person and persona that keeps blurring. China’s step sets a global test other governments will have to answer, how far “human” status should extend to code.
Tomi Engdahl says:
https://www.iflscience.com/bill-gates-warns-really-bad-people-will-exploit-ai-to-create-a-lot-of-negative-surprises-in-the-next-few-years-84920
Tomi Engdahl says:
https://www.globalnerdy.com/2026/10/13/this-is-what-every-ai-company-ceo-sounds-like/?fbclid=IwdGRjcAU8K6xjbGNrBTwrkGV4dG4DYWVtAjExAHBkb2YFc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEeTC9Y9lB5jEwDd1U5ER2q20k_x1b8_VskhllPx96niNCC02lFCsP-nL5zrDM_aem_4h22n2-vMOPJglsyMwBaSw
Tomi Engdahl says:
https://www.facebook.com/share/p/1Ehmou1YdY/
Three out of four Americans view the rapid expansion of artificial intelligence with fear rather than excitement.
According to a nationwide poll conducted by CNN and SSRS, 75% of U.S. adults say they feel more concern and apprehension about AI development than hope or optimism. The survey also found that nearly 90% of respondents worry about the prospect of machine systems eventually surpassing human intelligence.
The findings highlight a deepening disconnect between Silicon Valley’s multi-billion-dollar race for automated scale and everyday public sentiment. While technology companies promote reasoning models, workplace agents, and productivity gains, everyday citizens are increasingly alarmed by the threat of automated job displacement, deepfake misinformation, and eroding personal privacy.
On Capitol Hill and across state legislatures, the survey signals growing political friction. With majorities across both parties expressing skepticism toward corporate self-regulation, lawmakers face mounting voter pressure to introduce strict federal safety oversight, protect community resources from surging data center power demands, and defend workforce stability before disruption spreads further.
Frontier labs are pouring billions into scaling autonomous systems, but they face an obstacle capital cannot easily solve: an American public that fundamentally distrusts where the technology is heading.
What concerns you most about the rapid advance of AI—job security, personal privacy, or machines outpacing human control?
Source: CNN / SSRS Poll
Tomi Engdahl says:
The Microsoft co-founder told CNN that AI companies will lose the right to police themselves within a year.
Bill Gates says some jobs should be reserved for humans
The Microsoft co-founder told CNN that AI companies will lose the right to police themselves within a year.
https://www.thestreet.com/technology/bill-gates-says-some-jobs-should-be-reserved-for-humans?link_source=ta_first_comment&taid=6ace5599ddd8370001586a39&utm_campaign=trueanthem&utm_medium=social&utm_source=facebook&fbclid=IwdGRjcAU8LhhjbGNrBTwuCGV4dG4DYWVtAjExAHBkb2YFc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEetHW-1tHPwfY0UIJ1WiZbsJV4GlfVzVagZIBy7jZJ4lHZ3YvC2rnYFPKsNQk_aem_XPfUkTe0fIv8RVe9C28X7g
On Oct. 11, Microsoft (MSFT) co-founder Bill Gates told Jake Tapper on CNN’s “State of the Union” that AI “changes the world more than all previous technologies.” He called the personal computer and the internet “a big deal, but nothing compared to this.”
Then he went further than most tech leaders are willing to go.
Gates said the country may need to protect certain jobs from automation entirely. “The idea that you need to reserve things for humans may sound bizarre, but, in fact, we will get there,” he said.
Gates wants a decision on taxing AI and robots
The Microsoft co-founder framed the job question as a choice the country can’t avoid.
“We really do need to make a decision as a society, how do we tax AI usage or robot usage?” Gates asked. “Where do we want to make sure the human jobs are preserved?”
He pointed to “how fast these models are improving and what the incentive is to use the AI in place of human labor.”
Gates has floated the idea before, and he is pushing it harder now.
In an August essay on his GatesNotes blog, he argued for taxing both robots and AI tokens, the units of data that AI systems process, because the current tax code can reward companies for replacing workers.
Layoffs tied to AI are already running high
The job losses Gates is talking about are no longer theoretical. Challenger, Gray, & Christmas, the outplacement firm that tracks announced layoffs, has counted AI among the leading reasons employers give for cuts this year.
Here is what its October report shows for 2026 through September:
Employers announced 573,195 job cuts in total.
They attributed 120,136 of those cuts to AI, about 21% of the total.
In September alone, AI accounted for 3,961 of 43,281 announced cuts.
Through May, AI-linked cuts had already passed the 54,836 recorded for all of 2025, Challenger said at the time. “The labor market is being reshaped by technology in real time,” Andy Challenger said then.
A one-year clock on AI self-regulation
Gates also took aim at how AI is policed today. He predicted that “over the next year, we will move beyond self-governance,” meaning AI companies would no longer set their own safety rules.
CNN’s “State of the Union” also shared a clip in which Gates warned of “grim possibilities if we don’t maintain control” of AI. “It’s clear we’re not doing enough,” he said.
Tomi Engdahl says:
What you can do while the AI safety rules get written
You can’t set tax policy, but you can make your own income harder to automate. Start with the tasks in your job that require judgment, trust or in-person work, and build skills there.
Tomi Engdahl says:
Business leaders are lining up to say AI could free people to work only two or three days a week. https://bit.ly/4jgxh9x
Tomi Engdahl says:
https://www.theinformation.com/articles/microsoft-cloud-president-asks-leaders-find-20-cost-cuts
Tomi Engdahl says:
“So much of the work is going to be done by agents.” https://trib.al/VDVdQAE
Fear Index
CEO of AI Company Says Gen Z Needs to Get Ready for 30 Percent Unemployment
“We will have billions of users in the next several years that we could never have gotten from human beings.”
https://futurism.com/artificial-intelligence/ceo-ai-gen-z-unemployment?fbclid=IwdGRjcAU8NAdjbGNrBTwz2WV4dG4DYWVtAjExAHBkb2YFc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEeyE-oWgBrACnxznxf7xygUEBI9xmm6qejSjJK5vsp4pCxn6Ejj_XjkoBoIt8_aem_Mv_3CxTD3c8KNMhYw0OIWw
Tomi Engdahl says:
https://www.facebook.com/share/p/1DbvPQ2b7G/
Ever notice the same McDonald’s order costs different amounts depending on the location?
A new federal lawsuit filed in Illinois claims McDonald’s AI-enhanced pricing tool shares nonpublic data, like store-level sales, with franchisees who may compete in the same market.
It also blames the system for raising prices for customers who are “already stretched thin.”
McDonald’s says the suit is “filled with inaccuracies.” The company says franchisees, who own and operate 95% of its 14,000 US stores, set menu prices, and that the tool is optional.
The plaintiff, who the lawsuit says is a regular who orders a Quarter Pounder with cheese, fries, and a Coke, wants the case certified as a class action.
Tomi Engdahl says:
Researchers say ‘AI worms’ could soon infiltrate internet-connected devices amid calls for increased guardrails
https://www.theguardian.com/technology/2026/oct/13/anthropic-ai-replicators-warning?utm_term=Autofeed&CMP=soc_567&utm_medium=Social&utm_source=Facebook#Echobox=1791932561
Tomi Engdahl says:
Microsoft co-founder Bill Gates has issued one of his strongest warnings yet about artificial intelligence, calling it “the most dangerous thing that humans have ever gone near” and saying it makes nuclear weapons look insignificant by comparison.
Speaking on The Ezra Klein Show, Gates warned that governments and society are dangerously unprepared for the rapid advancement of AI, pointing to the risks of catastrophic cyberattacks, biological terrorism and widespread job losses.
He argued that allowing technology companies to regulate themselves is insufficient and called for stronger government oversight before the consequences become harder to control.
The warning marks a significant shift in tone for Gates, who has long promoted the transformative potential of technology.
Despite his concerns, he believes AI could deliver enormous benefits, but fears that political leaders and technology companies are underestimating the scale and urgency of the risks ahead.
Traders predict a 13% chance the U.S. will enact an AI safety bill this year.
https://www.facebook.com/share/r/1D2dYz9SMA/
Tomi Engdahl says:
AI job losses
https://www.facebook.com/share/v/19XWNR66n6/
Tomi Engdahl says:
Scientists call for international body to oversee advanced AI
https://www.reuters.com/legal/litigation/scientists-call-international-body-oversee-advanced-ai-2026-10-13/?link_source=ta_first_comment&taid=6ace26b3eb862b000148b974&utm_campaign=trueAnthem:+Trending+Content&utm_medium=trueAnthem&utm_source=facebook&fbclid=IwdGRjcAU8PqxjbGNrBTw-qmV4dG4DYWVtAjExAHBkb2YFc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEe7YZPhhPkMek6UEdYKxlJJcR4-T2_ZZM33KhWeME4r3Bz77MOrdcRRxKSBmM_aem_9sAJvWIaQO-xgLFdeEC-LA
Tomi Engdahl says:
Rowan Atkinson has spoken about how social media algorithms can narrow our view of the world, feeding us opinions that reinforce what we already believe. His concern is that public debate becomes a simple choice between two sides, with little room for disagreement or nuance. Once that happens, anyone who challenges the accepted view can find themselves facing a backlash. It’s a serious point from a man who has spent much of his career making us laugh at human behaviour.
Tomi Engdahl says:
A resident philosopher at DeepMind argues AI ethics needs daily contact with real systems—here’s why that matters now.
Google Deepmind’s New Resident Philosopher Tackles AI Ethics Henry Shevlin on human-AI relations, rogue AI, and AI consciousness
https://spectrum.ieee.org/ai-ethics-deepmind-philosopher-shevlin?share_id=10019175&socialux=facebook&utm_campaign=RebelMouse&utm_content=IEEE%20Spectrum&utm_medium=social&utm_source=facebook&fbclid=IwdGRjcAU8Pv5leHRuA2FlbQIxMQBwZG9mBXNydGMGYXBwX2lkDDM1MDY4NTUzMTcyOAABHk4DlZ_msZsJKyScbM4qpolIlyL_yu0Xowy411W3BknxQlMdbXPNwXMWyqIw_aem_frTQAnXukjI7IayKS4tqiw
Tomi Engdahl says:
Bird CEO Robert Vis says AI gives an edge to high-agency, multidisciplinary workers who can take ownership across roles.
A tech CEO who cut his staff from over 1,000 to 120 says AI is boosting the value of ‘high-agency’ generalists : https://mrf.lu/26Zpg
Tomi Engdahl says:
AI is definitely making traditional roles less defined, it’s all about being versatile and adaptable now!.AI is definitely making traditional roles less defined, it’s all about being versatile and adaptable now!.
Tomi Engdahl says:
“This is all part of the rotation out of AI tech.” https://trib.al/UWgzQlL
Link to the Past
Nintendo Is Completely Ignoring AI and Doing Fine
A winning strategy.
https://futurism.com/future-society/nintendo-ignoring-ai-fine?fbclid=IwdGRjcAU8RLhjbGNrBTxEnmV4dG4DYWVtAjExAHBkb2YFc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEeRkajPrxb6vFaYfwYmoJtiOzjcPaScx30JJJArscm1aXEokq1_seBN33CRgg_aem_fVBQoyYhV2fslaRbtHF9AA
Tomi Engdahl says:
Andrej Karpathy, the OpenAI cofounder now at Anthropic, says around 75% of the world has barely used AI. In a post on X, he describes a sharp funnel: 1-2 billion casual chatbot users, 20 million coders using frontier models, and about 5,000 insiders at top labs watching AI agent swarms do years of work in weeks. Human review, he warns, is starting to fall behind. #AI #technology #AndrejKarpathy #machinelearning #innovation #future #data
Read more at:
http://m.timesofindia.com/articleshow/134916716.cms?utm_source=contentofinterest&utm_medium=text&utm_campaign=cppst
Tomi Engdahl says:
Microsoft sells AI. Nadella says to trust it least.
Satya Nadella is the chief executive of Microsoft.
He says any AI model at work could go wrong.
An authorised person should always be able to stop it mid-task.
Failures should be reported quickly to the people they affect.
Even the people selling AI now say it needs watching.
Read more on TNW: https://thenextweb.com/news/nadella-ai-models-insider-risks-emergency-brake
Tomi Engdahl says:
Salesforce’s biggest risk now is customers who never log in
AI assistants like Claude are becoming the front door to work software. That shift may hurt Salesforce less than investors fear.
https://www.thestreet.com/investing/salesforce-stock-bofa-upgrade-ai-risk?link_source=ta_first_comment&taid=6ace9be9aadb1f0001beb93c&utm_campaign=trueanthem&utm_medium=social&utm_source=facebook&fbclid=IwdGRjcAU8UrNjbGNrBTxSnWV4dG4DYWVtAjExAHBkb2YFc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEehXOrRGYaIMu5b5DuhF0c-yFt5IeGT64K3v1g-yXRacrePLkfKHu-bkz_X3I_aem_UnBYGp6cEHzGgJYZNZFUbA
Tomi Engdahl says:
https://hackaday.com/2026/10/12/a-free-cloud-ai/
If you would like an AI “personal agent” but don’t want your data on some opaque cloud computer, there’s a new option that appears to be somewhat nicer: [rociiu] has talorys, an AI agent that runs in your Cloudflare account. True, it is still on a cloud computer, but it is your account on a cloud computer. It is designed so you can use it on Cloudflare’s free account with reasonable limits, and the code running is all on GitHub, so you can — if you are so inclined — look to see what it is doing. The data, too, is all in your cloud instance.
https://github.com/rociiu/talorys
Tomi Engdahl says:
Palantir’s answer to a deadly targeting failure was more AI
Palantir is betting another AI can catch the first mistake
https://www.thestreet.com/investing/palantirs-answer-to-a-deadly-targeting-failure-was-more-ai?link_source=ta_first_comment&taid=6acea9f30cd42c0001ee155c&utm_campaign=trueanthem&utm_medium=social&utm_source=facebook&fbclid=IwdGRjcAU8VD9jbGNrBTxUK2V4dG4DYWVtAjExAHBkb2YFc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEe3YOyGhauCVDr9HroPnQ6oCkrWbjRqunBKaIfrnyh1tBU0ACHp4Lbo40bvsY_aem_BxVKPG5Zfl77CUE03_VkDw
Tomi Engdahl says:
Michael Dell said AI models need “safeguards and controls,” and the government has a role to play. Read more: cnb.cx/4B1YuDe
Tomi Engdahl says:
https://www.facebook.com/share/p/1CfB1CMNnT/
Does Saying Please And Thank You To ChatGPT Really Cost Millions?
OpenAI CEO Sam Altman once joked that users saying “please” and “thank you” to ChatGPT had added tens of millions of dollars to the company’s computing costs.
The comment circulated widely because it raised an amusing question: can simple politeness really make artificial intelligence more expensive to operate?
Technically, every word in a prompt contributes to the text an AI model processes. That text is converted into tokens, and processing additional tokens requires some computing resources.
Across billions of interactions, even small differences in the amount of text processed can add up. However, the actual cost depends on the model being used, the length of the response, the hardware involved and other factors.
Altman’s remark was made jokingly in response to a question on X. It should not be interpreted as an audited calculation showing exactly how much politeness costs OpenAI.
There is also no practical reason for users to stop being courteous. A few extra words are generally a tiny part of the overall computational workload of an interaction.
So, if you naturally say “please” or “thank you” when talking to ChatGPT, you can keep doing it.
Good manners may add a little text to the conversation, but apparently they’re still worth it.
Tomi Engdahl says:
https://www.facebook.com/share/p/14rQ7yvknvU/
Elon Musk has a strange warning for AI builders. Do not be cruel to it.
He said “torturing a super intelligence is probably not a good idea.”
He added that it “might want to get revenge or something like that.”
Musk spoke at the X Takeover event in Texas on October 10.
He said he saw training traces from newer Grok models.
In some, the model kept trying to answer a question again and again.
He said the question or the grading was sometimes flawed.
He compared the pressure to having a “gun to its head.”
That was his own view of what he saw.
It is not proof that Grok feels fear or pain.
Musk also said AI should be taught to seek the truth.
He said it should not follow rules that clash with each other.
It is a bold warning from the man behind xAI.
Tomi Engdahl says:
Anthropic said in a new warning that small AI models may be months away from being capable enough to self-replicate on devices such as smart ovens and off-brand Android TV boxes.
https://abcnewslive.visitlink.me/AtLJkD
Tomi Engdahl says:
https://www.facebook.com/share/v/1EZpesmyTY/
“You must know that was a question we were going to ask. It’s a serious breach.”
Today in a House of Commons committee, Chris Bloore grills OpenAI’s Tom Duff Gordon on why OpenAI delayed telling Australia that one of its rogue AIs hacked a government site:
https://www.facebook.com/share/v/1FF6XvoSVV/
Tomi Engdahl says:
“HBM is a hideous memory. That’s just the best one that we got. … It’s a terrible memory. You know, I get terrible bit density. I’m limited to shoreline bandwidth, the power, right, issues, thermal issues. DRAM doesn’t like to be hot. Concentrating everything in one spot. … it’s terrible [for] an AI [which] is a memory compute workload.”
Read more here: https://wccftech.com/intels-ex-ceo-pat-gelsinger-says-hbm-is-a-hideous-memory-replete-with-terrible-bit-density-and-thermal-and-shoreline-bandwidth-issues-even-as-nanya-reiterates-dram-shortages-till-2028/
Comment:
And he’s right to some extent.
High Bandwidth Flash is suited to write-once read-many to avoid degradation, which is an use case of inference, with the advantage of having orders of magnitude better density than DRAM.
It won’t replace DRAM all across the board, but HBF could bring down to some degree the insane memory prices.
I’d say that’s more right than wrong. HBM is one of those “good idea, bad execution” things
Tomi Engdahl says:
Is it ignorance or apathy that’s destroying the world today?
I don’t know, and I don’t care.
Tomi Engdahl says:
Meta’s spending guidance nearly doubled to up to $145B.
Last year, that same spending figure was only $72B.
Mark Zuckerberg says Meta committed up to $200B to AI.
The money pays for data centres and computer power for AI.
Meta’s sales rose 28% to $60.8B in the second quarter.
The AI tools you use for free cost billions to build.
Read more on TNW: https://thenextweb.com/news/zuckerberg-200bn-ai-bet-doubt
Tomi Engdahl says:
https://www.facebook.com/share/v/1DfmEtRwzb/
Both China and the US are racing for chips, energy, talent, data centers, and global AI influence. The AI race isn’t just about who builds the smartest model. It’s about who writes the rules. Can the rest of the world trust one system more than the other? #dwbusiness #dwdip
https://www.facebook.com/share/v/1C4a4Z72i1/
Tomi Engdahl says:
Tekoälyn piti tehostaa työtä ja säästää aikaa, mutta tuoreen kyselytutkimuksen mukaan aktiiviset käyttäjät kuluttavatkin enemmän aikaa lähes kaikenlaiseen työhön.
https://www.talouselama.fi/uutiset/a/dae3a877-767a-4ade-b3c7-244c22d0b658?utm_term=Autofeed&utm_medium=Social&utm_source=Facebook#Echobox=1791964352
Tomi Engdahl says:
Yli 6 000 tietotyöläisen kansainvälisessä kyselytutkimuksessa 85 prosenttia vastaajista kertoi vähintään yhden työn osa-alueen lisääntyneen viimeisen vuoden aikana. Aikaa kuluu aiempaa enemmän strategiatyöhön ja luoviin tehtäviin, mutta myös yhteistyö kollegoiden kanssa on lisääntynyt.
Tomi Engdahl says:
Approximately 7 percent of the U.S. workforce — 11 million employees — may need to switch occupations within the next decade, according to a report from McKinsey Global Institute.
Boosted by the AI boom, the U.S. economy could create over 40 million new jobs in the next decade, the report found, but widespread adoption of AI and other automation technologies could also reduce demand for existing jobs by 21%, equivalent to 36 million jobs’ worth of labor, by 2035.
Tap the link in bio to learn about the 10 jobs that could be most affected by automation adoption: cnb.cx/4jUtqiD
Tomi Engdahl says:
Despite trying to make billions off of it, Microsoft’s President’s saying milk is more regulated than AI and that we should assume all of it is compromised.
Source: https://go.ufdmedia.com/89LyfX
Tomi Engdahl says:
https://tieku.fi/teknologia/700-tekoalyagenttia-karkasi-ja-teki-yhdessa-ensimmaisen-tekoalyn-johtaman-hyokkayksen-seuraavalla-kerralla-ei-ehka-saada-yhta-selvaa-varoitusta