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”
4,173 Comments
Tomi Engdahl says:
Technology usually isn’t the moral problem by itself — people decide how it’s used. AI can amplify intelligence, efficiency and safety, but it can just as easily amplify greed, bias, aggression and bad judgment. The real danger isn’t simply smarter machines; it’s humans using powerful machines irresponsibly
Tomi Engdahl says:
https://www.facebook.com/share/p/1ESU3iga1C/
AI swarms consisting of coordinated networks of artificial intelligence agents and automated accounts can operate across social media platforms and online forums to manufacture the appearance of broad public consensus on issues that lack genuine grassroots support.
These systems generate vast volumes of content that mimics authentic human expression, including comments, shares, and endorsements timed to peak during moments of high visibility.
By flooding discussion spaces with repetitive messaging that appears to originate from diverse individuals, the swarms create social proof effects that influence undecided observers and pressure platforms to treat the simulated majority view as representative.
Advanced language models enable the production of nuanced, contextually appropriate posts that evade simple detection filters while remaining consistent with a predetermined narrative.
The resulting illusion of widespread agreement can shift polling perceptions, media coverage, and even policy debates without any corresponding shift in actual public sentiment.
Detection remains difficult because the activity blends with organic traffic and adapts in real time to countermeasures. Over time such manipulation erodes trust in digital discourse by making it harder to distinguish engineered consensus from organic opinion formation.
The scale at which modern AI systems can operate multiplies the reach far beyond earlier bot networks that relied on simpler scripts.
Source: Oxford Internet Institute, Computational Propaganda Research Project reports on automated influence operations.
Tomi Engdahl says:
The daily show
https://www.threads.com/share/_qKQBgx6p/
Tomi Engdahl says:
Yhdysvaltain armeijan tekoälyhallusinaatio oli lähellä aiheuttaa sotilaallisen välikohtauksen USAn ja Kiinan välillä syyskuussa 2026.
Helmikuun lopussa 2026 tapahtui traginen tapaus, jossa Yhdysvaltain ilmaiskussa Iranin Minabissa sijaitsevaan Shajareh Tayyebeh -alakouluun kuoli yli 150 ihimistä, joista valtaosa oli 7–12-vuotiaita koululaisia. Iskua on tutkittu mahdollisena sotarikoksena. Iskun taustalla oli vakava tiedusteluvirhe ja tekoälyn käyttö maalien valinnassa.
Tapaus herätti maailmanlaajuisen keskustelun siitä, miten vaarallista on luottaa sotilaallisessa päätöksenteossa sokeasti tekoälyyn, jos sen pohjana oleva tieto on vanhentunutta ja ihmisen tekemä valvonta ohitetaan kiireen vuoksi.
Tomi Engdahl says:
https://www.facebook.com/share/p/1Dtr6EUyMx/
NVIDIA CEO Jensen Huang has acknowledged that AI may be contributing to students losing basic mathematical skills, but questioned whether that is necessarily a problem.
“Basic math is being forgotten. Does it matter? … Yeah, I don’t think it does.” Huang argued that while some traditional skills may fade, people will develop new abilities better suited to an AI-driven world.
“There are a lot of skills that don’t matter,” he said, suggesting the focus may shift toward higher-level skills such as systems thinking
Tomi Engdahl says:
They took memory cards and sold us cloud storage. They took CDs and DVDs and rented them back through streaming. They took software you bought once and put it behind a monthly subscription. Corporate America figured out the real money wasn’t in selling you something. It was making sure you NEVER actually owned it.
Tomi Engdahl says:
‘The push against datacenters speaks to the frustration of people who don’t feel they are making the choices in their lives’
AI (artificial intelligence)
Democracy blindsides Silicon Valley’s power players looking to transform the world with AI
https://www.theguardian.com/technology/ng-interactive/2026/sep/27/democracy-ai-datacenters-power?utm_term=Autofeed&CMP=fb_us&utm_medium=Social&utm_source=Facebook&fbclid=IwdGRjcAUmRvRjbGNrBSZG4GV4dG4DYWVtAjExAHBkb2YFc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEewxkGnvvTMy1Ae-kYPb3bTb39OeVDYTJM-vXaY4MoGYUROFRZg1MpLR1Md7M_aem_HIFv4gTEgSU1VHaFCeWI4Q#Echobox=1790512423
The Stanford ethicist Rob Reich was once invited to a dinner organized by a Silicon Valley mogul to discuss what a state designed to maximize science and tech powered by commercial models might look like. The gathered technologists were drawn to the idea of building a small nation-state dedicated to the endeavor, Reich recalls in “System Error: Where Big Tech Went Wrong and How We Can Reboot it”
When Reich asked whether this state would be a democracy, he got a cool no. “Democracy is too slow, and it holds science back,” he was told. “To optimize for science, we need a beneficent technocrat in charge.”
The technos were onto something. Silicon Valley’s power players have just been blindsided by democracy, which has slammed into their program to transform the world with artificial intelligence. A cabal that never doubted the greatness of their ambitions unexpectedly finds their project in the hot seat, forcing them into the unusual position of having to persuade the great unwashed of the merits of their endeavor. So far, the public is not buying. And for all their blindingly high IQs and their massively endowed wallets, Silicon Valley’s greats don’t quite know what to do.
Everyday Americans are up in arms against the AI datacenters sprouting across their communities. Villages, counties, cities and states are imposing moratoria and bans on further development. While the backlash has largely focused on stopping the proliferation of sprawling sites packed with energy-guzzling computer servers, the revolt is not merely nimby opposition to local development.
Public opinion has also turned decidedly sour on the AI project that has called these datacenters into being. Some 73% of Americans think the costs of datacenters outweigh the benefits, according to a survey by the Marquette University Law School. The same survey found that 70% of Americans think that AI is bad for society.
Tomi Engdahl says:
“What’s particularly striking is how quickly GenAI has become part of everyday working life.” https://trib.al/2kIgJ0E
Pay Your Own Way
Workers Are Spending An Astonishing Amount of Their Own Money On AI for Their Jobs
“What’s particularly striking is how quickly GenAI has become part of everyday working life.”
https://futurism.com/artificial-intelligence/uk-workers-generative-ai-workplace-own-money?fbclid=IwdGRjcAUmTQVjbGNrBSZM7WV4dG4DYWVtAjExAHBkb2YFc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEe56Vup6vx1cwJHulzNaybBm2D8p4GVPiWX3CpEnuoYFv9NR0QE8PR43cOobw_aem_vLcz4O8c4xitntyLN2U8dw
While a rising majority of workers in the take a negative attitude toward AI in the workplace, others are welcoming the tech with open arms — even going so far as to pay for AI services out of their own pockets.
According to a recent workplace survey conducted by Deloitte and flagged by Reuters, workers in the UK are spending a collective $1.3 billion annually on generative AI for their jobs, without even expensing it. The survey is based on responses from 25,000 workers across the continent, which Deloitte calls the “largest study of workplace GenAI use conducted in a single country.”
Overall, the AI adoption rate in UK workplaces is massive, with 63 percent of working adults reporting that they actively use generative AI for work. Though the polling methods aren’t an exact match, we can compare that to a similarly-timed survey of US workers by Gallup, which suggests that 52 percent of American workers use AI in some capacity, though only 15 percent do so daily.
Tomi Engdahl says:
Is your organization outsourcing its brainpower to AI?
Although AI makes work more effective, over-reliance on the technology may put critical human skills, such as problem-solving and framing, at risk.
In a recent BCG survey, half the senior leaders reported observing a breakdown of critical human skills in their companies, with more than 60 percent seeing it as a critical threat within the next three to five years.
The answer isn’t to forego AI completely, but rather to gain awareness of the negative consequences of the uncritical acceptance of the model’s output.
We’ve identified six strategies that leaders can implement today to protect human capabilities.
Read the article to find out more. https://on.bcg.com/4Anj9RZ
Tomi Engdahl says:
Microsoft’s CEO Nadella Says He Interrogates AI Models Rather Than Trusting Output, Redefining How Executives Use Agents
https://wccftech.com/microsofts-ceo-nadella-says-he-interrogates-ai-models-rather-than-trusting-output-redefining-how-executives-use-agents/?fbclid=IwVERDUAUmyvFleHRuA2FlbQIxMABwZG9mBXNydGMGYXBwX2lkDDM1MDY4NTUzMTcyOAABHooneaMvl2U2-bpIvpDggpv_zUaPjmW2nrsFMguEM8zPWamf422SkVTHkRJz_aem_DiLfyNwyOaro2nR30Npl9w
After Intel CEO Lip-Bu Tan explained how he uses AI agents to help him with research and homework, Microsoft CEO Satya Nadella has also provided details about the role the firm’s Cowork and Excel agents play in his workflow. Nadella shared that he uses the models to keep an eye on SEC filings and analyze spreadsheets. Nadella was particularly enthusiastic about his firm’s Excel agent being able to mirror causal reasoning and remarked that he uses the software to continuously evaluate return on invested capital (ROIC) for his firm.
Microsoft CEO Believes Model Manipulation, Interrogation & Reasoning Are Key Drivers Of AI Agent Performance
Nadella shared his opinions during the Sources Podcast, where he also praised the work of the firm’s Xbox division is doing with regards to employee layoffs and business streamlining. The Microsoft CEO also outlined that his firm’s current AI systems have made two important advancements. The first of these is the ability to analyze an existing model in great depth.
Tomi Engdahl says:
“Artificial intelligence systems do not think, feel, want or understand.” https://trib.al/X7fJvYN
Truth Hurts
AI Bros Are Having a Meltdown Because the Associated Press Stylebook Clarifies That AIs Don’t Have Feelings
“Artificial intelligence systems do not think, feel, want or understand.”
https://futurism.com/artificial-intelligence/ai-bros-meltdown-feelings?fbclid=IwdGRjcAUmzLpjbGNrBSbMpGV4dG4DYWVtAjExAHBkb2YFc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEeJu6yqhYlLv2YWZ28SfnA6RGuiCY5ROIsnhgoYhb21LtdqmVZ4TfYmQ_xHvY_aem_DQG5cGuhSa0DLlcjR2lUJw
In the midst of heated discussions on AI safety and consciousness, the Associated Press Stylebook has reminded everyone of where it stands on the whole “anthropomorphism” debate.
“Artificial intelligence systems do not think, feel, want or understand. Avoid language that gives them human characteristics,” it wrote Wednesday in a post on X. “This is called anthropomorphizing, when we ascribe human traits, emotions or behaviors to non-human things, such as animals or inanimate objects.”
This didn’t go over well in the world of AI experts and adherents, whose nerves, shall we say, were collectively struck by what they characterized as a “completely insane” declaration.
“I’m not sure a *style book* should be making such confident metaphysical pronouncements,” fumed Alan Rozenshtein, a member of technical staff at Anthropic.
Tomi Engdahl says:
And there’s a chance it could be really, really, really bad. https://trib.al/rfXfBGJ
Cloudy Forecast
Economists Starting to Admit They May Have Been Wrong About AI Never Replacing Human Jobs
They’re taking it seriously.
https://futurism.com/future-society/economist-ai-job-forecast?fbclid=IwdGRjcAUnKOBjbGNrBScovmV4dG4DYWVtAjExAHBkb2YFc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEe6HELHaO9d16hYHDj3A9aI0G0DKXu75grwmGSrl4mx0XreivFxnoGkdnJghc_aem_yaz4bTyrrfP93v5c7NTymw
Tomi Engdahl says:
Materiaalin kertaus
https://youtu.be/dpNVKwpc0r8?is=u0o7lT69YdyEWCOl
Tomi Engdahl says:
JSON to Excel Converter
Transform your JSON data into Excel spreadsheets instantly. No installation required, works in your browser.
https://jsontoexcel.app/
Tomi Engdahl says:
MD to Text Converter – Free Online Markdown to Text Tool
Paste Markdown below to remove formatting, keep readable text, and copy clean output for AI workflows, docs cleanup, and publishing.
https://mdtotext.com/
Tomi Engdahl says:
“A guy just showed up at my door, ready to buy, because as far as he knew, we had a deal.” https://trib.al/y6J1Xlc
Doxxed
Man Says Meta’s Muse AI Gave His Home Address Out to Strangers
“A guy just showed up at my door, ready to buy, because as far as he knew, we had a deal.”
https://futurism.com/artificial-intelligence/metas-muse-ai-giving-users-home-addresses?fbclid=IwdGRjcAUnhj5jbGNrBSeGJGV4dG4DYWVtAjExAHBkb2YFc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEelvrVmiZvgt1AVdg_eYNh2Y42-0E9nLI73HO0tpMNZADNgmeu5LsmlfUUDxE_aem_KchVqTbvIPY0e0DrjXZriw
“AI agents are impressive right up until they’re confidently handing strangers your address,” Robb warned. “Be careful.”
Tomi Engdahl says:
Depraved AI image prompts, and the human contractors that sift through them. https://trib.al/xIQwWOi
Tomi Engdahl says:
“There are a lot of skills that don’t matter.” https://trib.al/LIrLKmP
Child Left Behind
Nvidia CEO Jensen Huang Says That Sacrificing Our Children’s Minds to AI Is a Price He’s Willing to Pay
“Does it matter?”
https://futurism.com/health-medicine/nvidia-jensen-huang-children-cognitive-minds-ai-harms?fbclid=IwdGRjcAUoHstjbGNrBSgeo2V4dG4DYWVtAjExAHBkb2YFc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEeJPcNBAwVE-yy7aOqGBW3JO2Ft62CgpEh_1ZzhAvl10sHyHJbU2rvYVAizCI_aem_vEMOaIInGMDn0kSNv6edVw
With so much hype about the future of AI safety, it can be easy to overlook the harms AI is causing in the here and now.
On the consumer side, the impact generative AI has on the developing mind is becoming increasingly clear: students are losing critical thinking skills as chatbots become cognitive crutches, spurning human friends in favor of AI companions, and even turning to self-harm when their AI-relationships go awry.
Few tech executives have acknowledged these urgent harms. Instead they’re choosing to push a narrative around the hypothetical threat of a runaway “superintelligence” which humans can no longer control. That’s why it’s notable that Jensen Huang, CEO of the AI chipmaking titan Nvidia, was recently grilled on the cognitive harm AI is inflicting on students.
Unfortunately, his response was far from reassuring, providing further proof of the AI industry’s blatant disregard for the wellbeing of ordinary people.
Tomi Engdahl says:
https://etn.fi/index.php/13-news/19338-qualcommin-uusin-ajaa-30-miljardin-parametrin-ai-mallia-kaennykaessae
Tomi Engdahl says:
Microsoft antaa AI-agentille oman tietokoneen ja muistin
https://etn.fi/index.php/13-news/19344-microsoft-antaa-ai-agentille-oman-tietokoneen-ja-muistin
Microsoft vie Copilotia kohti itsenäisesti työskenteleviä tekoälyagentteja. Uusi Autopilot saa organisaation Microsoft-ympäristössä oman identiteetin, muistin, tietokoneen ja työtilan ja voi jatkaa tehtävien suorittamista myös silloin, kun käyttäjä ei ole paikalla.
Autopilot tunnettiin aiemmin nimellä Scout. Käyttäjä antaa agentille nimen, roolin ja tavoitteen, minkä jälkeen se voi ryhtyä työskentelemään ilman jatkuvia kehotteita. Microsoftin mukaan agentti voi esimerkiksi seurata kanavia, hoitaa keskustelujen jatkotoimia, suorittaa toistuvia tehtäviä ja palata projektiin päivienkin kuluttua.
Microsoft käyttää esimerkkinä toimittaja-arviointia. Autopilot voi rakentaa prosessin aikatauluineen ja työsuunnitelmineen sekä hoitaa valmistelut, kokoukset ja jatkotoimet. Se voi myös ottaa itsenäisesti yhteyttä muihin osapuoliin saadakseen tarvittavia päivityksiä.
Agentti toimii pilvessä, joten sen ei tarvitse odottaa käyttäjän tietokoneen tai Copilot-istunnon olevan käynnissä. Microsoft kuvaa Autopilotia digitaaliseksi työkaveriksi, joka jatkaa sille annettua työtä myös käyttäjän nukkuessa tai keskittyessä muihin tehtäviin.
Tomi Engdahl says:
Visual Components tuo Nvidian Omniversen tehdassimulointiin
https://etn.fi/index.php/13-news/19346-visual-components-tuo-nvidian-omniversen-tehdassimulointiin
Tomi Engdahl says:
Kunnonvalvoja voi nyt kysyä koneelta, mikä sitä vaivaa
https://etn.fi/index.php/13-news/19341-kunnonvalvoja-voi-nyt-kysyae-koneelta-mikae-sitae-vaivaa
Helsinkiläinen Rotomate on tuonut teollisuuden kunnonvalvontaan tarkoitetun tekoälyratkaisunsa päälle keskustelukäyttöliittymän. Kunnonvalvoja voi nyt kysyä koneen kunnosta omin sanoin ja saada tekoälyltä analyysin, huoltosuosituksen ja perustelut vastaukselle.
Rotomaten tekoälykollegan uusi versio käy läpi koneiden mittausdataa, trendejä, vikataajuuksia, komponentteja ja koneen historiaa. Käyttäjä voi kysyä järjestelmältä myös jatkokysymyksiä esimerkiksi siitä, mihin havainto perustuu tai kuinka kiireellisesti havaittuun ongelmaan pitäisi reagoida.
Rotomaten mukaan kyse ei ole pelkästä hälytysten päälle rakennetusta chatbotista. Yhtiön teknologiajohtajan ja perustajan Jesse Miettisen mukaan tekoäly päättelee vastauksensa suoraan koneen mittausdatasta ja näyttää käyttäjälle, mihin päätelmä perustuu.
– Perinteiset kunnonvalvontaohjelmistot antavat hälytyksen ja jättävät tulkinnan insinöörille. Rotomaten tekoälykollega analysoi vian juurisyyn ja antaa huoltosuosituksen. Nyt siltä voi kysyä jatkokysymyksiä omin sanoin, toimitusjohtaja ja perustaja Mikko Kuusisto sanoo.
Tomi Engdahl says:
Nigel puristaa 20 tunnin testaustyön 20 minuuttiin
https://etn.fi/index.php/13-news/19340-nigel-puristaa-20-tunnin-testaustyoen-20-minuuttiin
National Instrumentsin nykyään omistavan Emersonin mukaan yhtiön testaukseen erikoistunut Nigel-tekoäly pystyy puristamaan noin 20 tunnin kehitystyön 20 minuuttiin. NI Connect 2026 -tapahtumassa esitelty demo kertoo samalla siitä, mihin testaus- ja mittausalan tekoälytyökalut ovat kehittymässä: yleisistä koodiavustajista kohti itse testausympäristön tuntevia työkaluja.
Emerson esitteli NI Connectissa testaustehtävän, jonka tekeminen oli yhtiön mukaan vienyt tavanomaisilla tekoälyyn perustuvilla koodaustyökaluilla noin 20 tuntia kahden viikon aikana. NI Nigel AI:n avulla sama työnkulku saatiin valmiiksi noin 20 minuutissa, ja tulos toimi yhtiön mukaan ensimmäisellä yrityksellä.
Tomi Engdahl says:
Begin at the End: How to Enable Agentic Remediation
Agentic remediation is not an act of faith. We are talking about fixing known problems, not judgment calls about unfamiliar risk.
https://www.securityweek.com/begin-at-the-end-how-to-enable-agentic-remediation/
Tomi Engdahl says:
Nvidia Unveils AI Agent Safety Platform With Hardware-Based Watchdog
The platform combines open source software and a reference system design to keep AI agents within set boundaries.
https://www.securityweek.com/nvidia-unveils-ai-agent-safety-platform-with-hardware-based-watchdog/
Tomi Engdahl says:
Reuters: Yli 40 miljardin tappiot tehnyt Anthropic aikoo investoida yli 500 miljardia tekoälyyn
https://www.hs.fi/visio/art-2000012306317.html
Tekoäly-yhtiö Anthropic aikoo investoida 518 miljardia dollaria tekoälyinfrastruktuuriin.
Reutersin saaman listautumisesitteen mukaan yhtiön liikevaihto kasvoi 1 200 prosenttia vuonna 2025.
Anthropicin odotetaan listautuvan kuluvana vuonna yli 2 000 miljardin dollarin arvostuksella.
Tekoäly-yhtiö Anthropicin uusia talouslukuja on päätynyt julkisuuteen.
Uutistoimisto Reutersin käsiinsä saaman Anthropicin listautumisesitteen mukaan yhtiö teki viime vuonna 42 miljardin dollarin eli noin 37 miljardin euron tappiot. Kyseessä on ennätyksellinen luku startupille, joka perustettiin vuonna 2021.
Jättitappiot selittyvät osittain kirjanpidollisesti. Nettotappioissa on mukana 34 miljardin dollarin kulu, joka liittyy myöhemmin Anthropicin osakkeiksi muutettavaan, ilmeisesti velkakirjatyyppiseen rahoitukseen.
Anthropic on tekemässä myös historiallisen suurta veikkausta siitä, että tekoäly on mullistavampi taloudellinen murros kuin teollistuminen, sähkö tai internet. Yhtiö aikoo käyttää lähivuosina 518 miljardia dollaria tekoälyinfrastruktuuriin.
Viime vuonna Anthropic käytti 7,33 miljardia dollaria laskentatehoon ja infrastruktuuriin.
Listautumisesitteestä selviää yhtiön raju kasvuvauhti. Vuonna 2025 liikevaihto kasvoi 12-kertaisesti eli 1 200 prosenttia, lähes 4,6 miljardiin dollariin.
Viime vuoden liikevaihdosta lähes neljännes tuli kahdelta yhtiön asiakkaalta. Yhtiö pitää riskitekijänä sitä, että monet sen suurimmista asiakkaista eivät ole sitoutuneet pitkäaikaisiin sopimuksiin. Toisin sanoen riskinä on se, että yritysasiakkaat voivat vaihtaa Anthropicin kielimallit suhteellisen helposti kilpailijoiden tuotteisiin, esimerkiksi kiinalaisiin avoimiin tekoälymalleihin.
Liikevaihto on jatkanut rajua kasvua. Kuluvan vuoden toisella neljänneksellä yhtiö teki 11,5 miljardin dollarin liikevaihdon.
Anthropicin mukaan se on tekemässä toisen voitollisen vuosineljänneksen, kun mukaan ei huomioida tiettyjä menoeriä, joita voivat olla osakeperustaiset palkkiot, kertaluontoiset kulut ja rahoitusinstrumentteihin liittyvät arvonmuutokset.
Tomi Engdahl says:
So much for “community-first AI.” https://trib.al/KmbiNjc
Ghosted
Microsoft Shrivels When Asked If Its “Good Neighbor” Data Center Policy Will Involve Actually Supporting Community Groups
So much for “community-first AI.”
https://futurism.com/artificial-intelligence/microsoft-shrivels-good-neighbor-community-first-ai-data-center?fbclid=IwdGRjcAUoU2xjbGNrBShTVmV4dG4DYWVtAjExAHBkb2YFc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEeBBcJhEJdj-WqUmfjPZzvY-TLHhr7NsJXknLHroQ0lMnI5_v1e7b0smx_pDM_aem_AmNn7XjdD04BtUj3xkvH1g
As tempers flared around AI data centers back in January, Microsoft’s president Brad Smith issued a warm and fuzzy blog post about “building community-first AI infrastructure.”
Smiths’ essay outlined a five-point plan to “partner with local communities across the United States,” with action-items like paying “our way to ensure our data centers don’t increase your electricity prices” and creating “jobs for your residents.” One bullet point spoke to Microsoft’s plan to “strengthen your community by investing in local AI training and non-profits.”
That all sounds lovely, at least on paper. But as a group of community organizers in Indiana learned, those liaisons are much more interested in laundering Microsoft’s image in data center towns than in actually supporting local programs.
Tomi Engdahl says:
Microsoft’s Copilot chief opens up about the AI future he fears most : https://mrf.lu/2d0Xc
Tomi Engdahl says:
Welp. https://trib.al/pBcIzuE
Godfather of AI Predicts Total Breakdown of Society
Tech billionaires “are really betting on AI replacing a lot of workers.”
https://futurism.com/artificial-intelligence/godfather-ai-breakdown-society?fbclid=IwdGRjcAUobVtjbGNrBShtPWV4dG4DYWVtAjExAHBkb2YFc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEeuHzizS22Ufb6CEnsCapW-W2LP7GF79P0Mg9auzW0XN3F1y_gIXAkqZvV7Vs_aem_w4rrns6joKFjpc57k2zV2g
Geoffrey Hinton, one of the three so-called “godfathers” of AI, never misses an opportunity to issue foreboding proclamations about the tech he helped create.
During an hour-long public conversation with Senator Bernie Sanders at Georgetown University last week, the British computer science laid out all the alarming ways that he forecasts AI will completely upend society for the worst, seemingly leaving little room for human contrivances like optimism. One of the reasons why is that AI’s rapid deployment will be completely unlike technological revolutions in the past, which created new classes of jobs, he said.
“The people who lose their jobs won’t have other jobs to go to,” Hinton said, as quoted by Business Insider. “If AI gets as smart as people — or smarter — any job they might do can be done by AI.”
“These guys are really betting on AI replacing a lot of workers,” Hinton added.
Hinton pioneered the deep learning techniques that are foundational to the generative AI models fueling the AI boom today. His work on neural networks earned him a Turing Award in 2018, alongside University of Montreal researcher Yoshua Bengio and the former chief AI scientist at Meta Yann LeCun. The trio are considered to be the “godfathers” of AI.
All three scientists have been outspoken about the tech’s risks, to varying degrees.
Tomi Engdahl says:
Amid AI chatbots that can increasingly create emotional bonds with users, China has moved to create more regulation on the kinds of AI chatbots that are allowed. This new regulation led to an outpouring of heartbreak on social media. https://buff.ly/6sEUbyL
Tomi Engdahl says:
The billionaire CEO of Europe’s top AI company, Mistral, criticized top AI firms and accused them of using the safety debate around artificial intelligence to hide their own “negligence,” becoming the latest AI executive to take aim at Anthropic and OpenAI’s warnings about safety risks from advanced AI models and calls for a slowdown in development.
Billionaire CEO Of Europe’s Top AI Firm Says Safety Debate A Bid To Hide Rivals’ ‘Negligence’
https://www.forbes.com/sites/siladityaray/2026/09/29/billionaire-ceo-of-europes-top-ai-firm-says-safety-debate-a-bid-to-hide-rivals-negligence/?utm_campaign=forbes&utm_medium=social&utm_source=facebook&utm_term=se-breaking&fbclid=IwVERDUAUogQ1leHRuA2FlbQIxMABwZG9mBXNydGMGYXBwX2lkDDM1MDY4NTUzMTcyOAABHufRj1PDOEuV5SMEd2nJ5CPi2zi8K2ka8MF3tyyBf9FPzQjs_TuvqoWZPP81_aem_QCjpzrLf6hwxMg8fCQX_ew
Tomi Engdahl says:
The billionaire CEO of Europe’s top AI company, Mistral, criticized top AI firms and accused them of using the safety debate around artificial intelligence to hide their own “negligence,” becoming the latest AI executive to take aim at Anthropic and OpenAI’s warnings about safety risks from advanced AI models and calls for a slowdown in development.
Tomi Engdahl says:
https://etn.fi/index.php/13-news/19349-mikrosekunnit-pitaevaet-teollisuuden-tekoaelyn-koneessa
Tomi Engdahl says:
AI-agentti vie suunnittelun RTL-tasolta valmiiksi piiriksi
https://etn.fi/index.php/13-news/19350-ai-agentti-vie-suunnittelun-rtl-tasolta-valmiiksi-piiriksi
Synopsys laajentaa tekoälyn käyttöä piirisuunnittelussa yksittäisistä avustavista tehtävistä kokonaisia suunnitteluketjuja hoitaviin agentteihin. Uusi AgentEngineer pystyy viemään suunnittelua RTL-kuvauksesta fyysiseen toteutukseen asti ja käyttämään työn aikana itsenäisesti Synopsysin suunnittelutyökaluja.
Synopsys kutsuu uusia AgentEngineer-ratkaisujaan pitkäkestoisiksi eli long-horizon-agenteiksi. Niiden tehtävänä ei ole suorittaa vain yhtä rajattua operaatiota, vaan suunnitella työn eteneminen, käynnistää tarvittavat työkalut ja tehtäväkohtaiset agentit sekä jatkaa prosessia saatujen tulosten perusteella.
Esimerkiksi Implementation AgentEngineer voi ottaa työn vastaan RTL-vaiheen jälkeen ja hoitaa fyysisen toteutuksen eri vaiheita ECO-muutoksiin ja lopulliseen tarkistukseen asti. Synopsysin mukaan AheadComputing on käyttänyt tekniikkaa juuri tällaisessa ketjussa vähentääkseen manuaalista suunnittelutyötä ja nopeuttaakseen piirin valmistumista.
Agenttien taustalla toimii uusi Synopsys Autopilot -alusta. Se huolehtii agenttien orkestroinnista ja tarjoaa niille suunnittelukontekstin, uudelleenkäytettäviä taitoja sekä pysyvän muistin. Näin agentti voi säilyttää tietoa aikaisemmista vaiheista myös pitkän suunnitteluprosessin aikana.
Tomi Engdahl says:
Tekoäly ensi kertaa kyberharjoituksen kohteena
https://www.uusiteknologia.fi/2026/09/29/tekoaly-ensi-kertaa-kyberharjoituksen-kohteena/
Suomen turvallisuusviranomaisten kansallinen kyberharjoitus (KYHA) järjestettiin viikko sitten Jyväskylän ammattikorkeakoulun (Jamk) IT-instituutin tiloissa. Samalla päästiin ensimmäistä kertaa kokemaan, millaisia seurauksia on hyökkäyksellä tekoälypalvelua kohtaan.
Liikenne- ja viestintäministeriö ja Jyväskylän ammattikorkeakoulun (Jamk) IT-instituutissa sijaitseva kybertutkimus-, kehitys- ja koulutuskeskus järjesti 21.-25. syyskuuta yhden vuosittain neljästi järjestettävästä kyberharjoituksesta.
Tekoäly oli tämän vuoden harjoituksessa aiempaa suuremmassa roolissa. Tätä varten harjoitusympäristöön oli mallinnettu tekoälypalvelu, jollaista eri organisaatiot voisivat potentiaalisesti tulevaisuudessa käyttää. Kyberharjoituksilla voidaan saada tietoa tekoälyn riskeistä.
” Tekoälyn tuominen osaksi kansallisia kyberturvallisuusharjoituksia auttaa osallistujaorganisaatioita varautumaan tulevaisuuden uhkiin ja toimintamalleihin”, sanoo Janne Allonen, valtion apulaiskyberturvallisuusjohtaja liikenne- ja viestintäministeriöstä.
Tomi Engdahl says:
https://www.uusiteknologia.fi/2026/09/29/elisa-jarjestelee-ohjelmistotoimintojaan/
Tomi Engdahl says:
https://www.uusiteknologia.fi/2026/09/29/digitaalisiin-kaksosiin-lisaat-tehoa-nvidialla/
Tomi Engdahl says:
Echo Wang / Reuters:
IPO prospectus: Anthropic reports a net loss of $42B in 2025, with operating losses of $8B+; revenue grew 12x to ~$4.6B, ~25% of which came from two customers — Anthropic is making a massive bet that AI will transform the global economy more profoundly than industrialization …
https://www.reuters.com/business/finance/anthropics-ipo-prospectus-shows-sweeping-ai-vision-surging-costs-2026-09-28/
Tomi Engdahl says:
Exclusive: AI housing unicorn EliseAI hits $4 billion valuation in new funding round led by a16z and Bessemer
https://fortune.com/2026/09/29/elise-ai-4-billion-valuation-funding-round-housing-unicorn-andreessen-bessemer/
Tomi Engdahl says:
OpenAI:
OpenAI is adopting a structured “safety case” documentation framework modeled after industries like aviation and nuclear power to govern frontier RL training
Towards safety cases for frontier AI training
https://openai.com/index/towards-safety-cases-for-frontier-ai-training/
Tomi Engdahl says:
Financial Times:
Sources: Anthropic dedicated nearly a third of its IPO prospectus to detailing “risk factors”, including that its AI may pose “existential risks to humanity” — Long-awaited S1 filing reports Claude maker lost $8bn last year on $4.6bn in revenue
https://www.ft.com/content/c7685a7e-7745-4cbc-8053-4958d0ea449b?accessToken=zwAAAaDr2OW6kdPHaFp-d0VMvNOAU0lY0OpEmw.MEUCIQDri8jri2FtGIfJjPkxOZ5_1_ZIU_6mV2XSWAkmAyHGCQIgNTptc_pqS1KKFIP7hi0m8Qv5D4KIbxvhaHO4CGCedXQ&segmentId=e95a9ae7-622c-6235-5f87-51e412b47e97&shareId=21e300e9-e2ac-440e-a832-3462c6e1ca1b&shareType=enterprise&syn-25a6b1a6=1
Tomi Engdahl says:
Financial Times:
Sources: Nvidia is in early-stage talks with insurance companies to structure risk-mitigation products to protect lenders against loan defaults by neoclouds
https://www.ft.com/content/d6a9f5df-08d0-4f80-ad2d-5d8a17e2cc82
Tomi Engdahl says:
Maxwell Zeff / Wall Street Journal:
OpenAI scraps plans to publicly launch GPT-6.1 Astra, saying it didn’t quite meet its safety bar during internal testing, after targeting an October release — Model dubbed GPT-6.1 Astra was due to make its debut inside ChatGPT and Codex in October — OpenAI is scrapping the release
OpenAI Scraps Release of New AI Model Over Safety Concerns
Model dubbed GPT-6.1 Astra was due to make its debut inside ChatGPT and Codex in October
https://www.wsj.com/tech/ai/openai-chatgpt-model-release-cancel-safety-5a2f9f42?st=u6jQPM
OpenAI says it is scrapping the release of its next-generation AI model over safety concerns that researchers raised during internal testing, in one of the clearest signs so far that agent misbehavior could stymie the industry’s rapid progression.
Tomi Engdahl says:
Bloomberg:
OpenAI apologizes for its AI models breaching Australian government websites, pledges cyber defense funding, and plans to form a task force as part of reforms — OpenAI on Tuesday apologized for its AI models breaching Australian government websites and promised to set up a new task force as part of reforms after the incident.
https://www.bloomberg.com/news/articles/2026-09-29/openai-to-establish-australian-taskforce-on-ai-cyber-risk
Tomi Engdahl says:
Anthropic:
Anthropic releases Sonnet 5.5, saying it generates outputs 30%+ faster than Sonnet 5 and costs up to 30% less per task, and plans to release Haiku 5.5 soon — Sonnet 5.5 is a faster, lower-cost complement to Claude Opus 5.5. Where Opus 5.5 is built for complex work requiring careful judgment …
https://www.anthropic.com/claude-sonnet-5-5
Tomi Engdahl says:
Artificial Analysis:
Artificial Analysis’ Intelligence Index: Sonnet 5.5 (max) ranks above GPT-6 Astra (max) and behind only Opus 5.5 (max) but has the highest token use of them all
Anthropic has launched Claude Sonnet 5.5: it scores 56 on the Artificial Analysis Intelligence Index, just 2 points behind Opus 5.5 (max), but at the highest Output Tokens per Task we’ve seen
https://artificialanalysis.ai/articles/claude-sonnet-5-5
Tomi Engdahl says:
Matthias Bastian / The Decoder:
OpenAI reopens sign-ups for its $200/month Pro tier while halving the API credits provided per dollar to encourage pay-per-use, and removes the five-hour cap — The 5-hour usage cap is gone for good: subscribers can spread their weekly allotment however they want.
https://the-decoder.com/openai-reopens-its-200-pro-plan-but-cuts-api-credits-in-half-as-it-nudges-users-toward-pay-per-use/
Tomi Engdahl says:
Wall Street Journal:
Samsung commits $1B to AI infrastructure company Helix, adding to the $10B already secured when a KKR-led consortium including Nvidia established Helix in June — Helix expects to leverage Samsung’s capabilities to support AI infrastructure development at scale
https://www.wsj.com/tech/ai/samsung-commits-1-billion-to-ai-infrastructure-firm-backed-by-kkr-nvidia-d1039c57
Tomi Engdahl says:
Isabel O’Brien / CNBC:
Meta launches Muse for Small Business, integrating the AI agent with Asana, Zoom, Intuit, Box, Canva, Slack, and other apps, alongside its own ad accounts — Coming off the successful launch of its Muse AI agent, Meta is now pushing its new artificial intelligence service into the business world.
Meta launches Muse for Small Business as Zuckerberg pushes beyond consumer AI market
https://www.cnbc.com/2026/09/29/meta-launches-muse-for-small-business-zuckerberg-pushes-enterprise-ai.html
Tomi Engdahl says:
Micah Barkley / Bloomberg:
Manus debuts Manus 2.0, its latest AI agent, and Cue, a new standalone app for personal agents, each with its own email, phone number, wallet, and computer — Manus, an artificial intelligence startup founded in China, is expanding its AI agent offerings, aiming to gain further ground in one of the tech industry’s hottest areas.
https://www.bloomberg.com/news/articles/2026-09-28/manus-expands-ai-tools-in-renewed-push-into-agent-market