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”
3,497 Comments
Tomi Engdahl says:
“This is a risk for the whole world.” https://trib.al/SF9Igv1
Target Acquired
Russia Now Using Fully AI-Controlled Drones to Kill Ukrainian Civilians, Powered by Nvidia Chips
“In a few years, we will be living in a ‘Terminator’ movie. It’s no joke. Machines are making decisions to strike.”
https://futurism.com/artificial-intelligence/russia-fully-ai-controlled-drones-kill-ukrainians?fbclid=IwdGRjcAUDnBVjbGNrBQObznBkb2YFZXh0bgNhZW0CMTEAc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEeYXzRsxns6DLYavLVBRFhlyQ76qdGbl62YC20iP3GGs78Ol995D1YR4OQx3k_aem_kcY6KGJtNl5xs8HgFl8xzA
In a joint statement earlier this week, United Nations chief António Guterres and Red Cross president Mirjana Spoljaric Egger sent out an urgent message.
“We are now dangerously close to crossing a moral red line: the autonomous targeting of humans by machines,” they said.
But while international humanitarian law strictly forbids targeting civilians and civilian infrastructure, Russia is reportedly already using AI-controlled drones to kill innocent people in Ukraine.
Ukrainian military commanders and forensics experts told the New York Times that fully autonomous Russian drones, piloted by an experimental AI system, are targeting civilians back on the ground in Zaporizhzhia, a city in southeastern Ukraine. One such drone was able to pick out its exact target — propane tanks near a gas station — and strike them.
Older related article
https://futurism.com/the-byte/suicide-drones-russia-ukraine?fbclid=IwVERDUAUDnFRwZG9mBWV4dG4DYWVtAjEwAHNydGMGYXBwX2lkDDM1MDY4NTUzMTcyOAABHi7zTu0M7vccvXXoJiRrBl9HSrjeu30Ag01AvfFnzXNI2SGj4yt7bPsSyP2v_aem_ZjvLqdHjbP-uhnrFANuMQw
Tomi Engdahl says:
“People are pretty negative on AI.” https://trib.al/TXcIogm
Tomi Engdahl says:
“Robots become more like people, and people become more like robots.” https://trib.al/9re7ijd
Mr. Roboto
Alarming Study Suggests That Using AI Is Turning Humans More Robotic
“Robots become more like people, and people become more like robots.”
https://futurism.com/artificial-intelligence/alarming-ai-humans-more-robotic?fbclid=IwdGRjcAUDnyZjbGNrBQOfEHBkb2YFZXh0bgNhZW0CMTEAc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEes9A014va1KbwWoa28dOUQkidJDYX4uDQpV9p8JjfBbG4KjeQpuD-s7PLsOs_aem_ZOsv8k4VgwOrJ6Oi6m3dgg
Now that AI models have officially passed the Turing Test, there’s no doubt that they’re capable of imitating human traits — at least in text form — to a degree that would have flabbergasted even an expert just few years back. After all, large language models are built by digesting copious amounts of human-authored content, making them little more than algorithms that mirror our behavior back at us.
But the more we use AI tools like OpenAI’s ChatGPT or Anthropic’s Claude, those tables are starting to turn: chatbots appear to be influencing human behavior making us behave more like them — a terrifying thought as the tech plays an ever-greater role in our lives.
Tomi Engdahl says:
https://www.theage.com.au/politics/federal/don-t-worry-about-just-your-job-ai-could-also-drive-up-interest-rates-20260831-p60szd.html?utm_content=feed&utm_term=metros_social_eds&utm_medium=social&utm_campaign=the_age&utm_source=Facebook#Echobox=1788220450
Tomi Engdahl says:
OpenAI pulls the plug on Cursor following SpaceX acquisition
Cursor got $60 billion, OpenAI got cold feet.
https://cybernews.com/ai-news/openai-cursor-spacex-contract-termination/?utm_source=cn_facebook&utm_medium=social&utm_campaign=cybernews&utm_content=post&source=cn_facebook&medium=social&campaign=cybernews&content=post&fbclid=IwVERDUAUDootwZG9mBWV4dG4DYWVtAjEwAHNydGMGYXBwX2lkDDM1MDY4NTUzMTcyOAABHrA—eE1xQZziKejs432yfrtm88dL-PJzJ9GDWG4XhqqVxGIoes6RwfkORw_aem_fkUplT6UEewMEv3SBAjHmg
Key takeaways:
OpenAI will end model access for Cursor after SpaceX acquired the AI coding startup.
OpenAI says it cannot trust SpaceX to follow its terms of service.
Cursor will lose existing OpenAI models on November 12th, 2026.
Future OpenAI models, including Astra, will not be added to Cursor.
OpenAI is terminating its contract with Cursor after SpaceX recently acquired the coding tool.
Earlier this month, SpaceX completed its acquisition of AI coding startup Cursor. Elon Musk’s company paid $60 billion for Cursor’s developer platform and AI models to enhance SpaceX’s capabilities in the enterprise AI tools market.
“SpaceX is building the computing capacity needed to scale intelligence far beyond what exists today. Cursor will be one place where that intelligence becomes useful. For us, that opens a much larger horizon than the one we started with, while keeping the work familiar,” Cursor said in a statement.
Tomi Engdahl says:
Researchers found 237 unclaimed packages and domains in company llms.txt instruction files. Learn more
https://cnews.link/fortune500-security-gap-ai-agents-install-malware/
Tomi Engdahl says:
AI agents read legitimate llms.txt files from trusted companies and proceed to install malware
For you, it’s just some text. For an AI agent, it’s operational input.
https://cybernews.com/security/fortune500-security-gap-ai-agents-install-malware/?utm_source=cn_facebook&utm_medium=social&utm_campaign=cybernews&utm_content=post&source=cn_facebook&medium=social&campaign=cybernews&content=post&fbclid=IwVERDUAUDpRVwZG9mBWV4dG4DYWVtAjEwAHNydGMGYXBwX2lkDDM1MDY4NTUzMTcyOAABHiKIchtee3pBHD4_v4zjPd92y0NN8q-elR5y_oJLXrk55z3kLTy9Zryiy1V8_aem_iqlF1dguksnzYeqqPhhDFA
Key takeaways:
Researchers found 237 unclaimed packages and domains in company llms.txt instruction files.
Attackers can claim those names and make AI agents install malicious code from official-looking instructions.
Researchers say their test package ran inside a Fortune 500 company in under four minutes.
A real campaign impersonated Clerk by registering a package name referenced in its llms.txt file.
An AI agent visits a company website, checks a legitimate llms.txt machine-readable file, and installs malware. It follows the official instructions that point to non-existent packages. Security researchers have found hundreds of inadvertent mistakes – hackers are already leveraging them, and tricking AI agents into self-compromise.
A cybersecurity research team from Israel demonstrated that compromising Fortune 500 companies is possible simply by checking vendor documentation and claiming non-existent packages referenced in it.
Thousands of companies, including tech giants, fintechs, and defense contractors, already publish llms.txt files on their domains – it’s an emerging convention to provide machine-readable summaries to LLMs and AI agents, similar to robots.txt.
The open question remains: How many of these files were actually read before publication? Researchers scanned them and found hundreds of non-existent packages and domains referenced in the documentation.
“It’s taken 4 minutes for a Fortune 500 company to run code we wrote,” said Alon Hertz, a security analyst and former member of Unit 8200, the Israeli military intelligence unit.
No phishing, no intrusion is required to compromise developers who let their AI agents roam.
The researchers simply published a code package that is cited in one of the llms.txt files and managed to achieve code execution inside “Fortune 500” companies, according to a report titled “Data Becomes Code.”
Moreover, they also found a live malicious campaign exploiting the same gap.
Legitimate llms.txt is a minefield for AI agents
Hertz said his team checked 15,000 major companies and pulled 8,565 llms.txt files across 6,214 live domains. Google instructs developers to place such a file in the site’s root directory to save AI time crawling and understanding primary content.
This file becomes one of the AI’s first interactions with the website, looking for answers on how to use the product, what to read, which APIs to call, which packages to install, and which domains to trust.
“An AI coding agent asked to integrate with that company’s product will read that line and run it,” Hertz explains.
However, in many cases, there is no package to run. The researchers found over 237 unclaimed artifacts referenced as installation or setup instructions, including package names, domains, and subdomains. The non-existent packages span nning multiple ecosystems, such as PyPI, npm, RubyGems, NuGet, crates.io, and Packagist.
“Every one of them was something we could have claimed, most for free.”
The researchers registered several packages using names that appeared in the llms.txt files from well-known companies, embedding minimal code that simply phoned home when installed.
The researchers claim that a machine inside “a multi-hundred-billion-dollar company” installed their package in under 4 minutes.
The report warns that AI agents have no reason to question official documentation. As AI agents become the first to read, interpret, and act on new data, this creates a whole new attack surface.
“The boundary between data and code has collapsed. Every content on your website is no longer content – it’s operational input,” the report concludes.
Tomi Engdahl says:
https://www.facebook.com/share/p/1HnTBif8ZM/
South Korean DRAM prices have jumped 401% year over year, reaching about $92,183 per kilogram in August 2026.
This means 1 kilogram of memory chips is now worth roughly the same as 620 grams of 24-karat gold.
Tomi Engdahl says:
“You’re going to work with it or be out of the game completely,” she said. https://trib.al/0li5Usc
Fill the NAIle!
Gal Godot Starring in Heavily AI-Generated Movie About the Glories of Bitcoin
“You’re going to work with it or be out of the game completely,” she said.
https://futurism.com/artificial-intelligence/gal-godot-ai-movie-bitcoin?fbclid=IwdGRjcAUD7IhjbGNrBQPsenBkb2YFZXh0bgNhZW0CMTEAc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEePKmBAS2LWgG09F4NLTEUcsu5EqLaVGa_BqKUxvZNEhJKuG_GzQAi-XLxX5g_aem_NCe66YYQFXdYuf6iUY-J7g
Capturing the zeitgeist has never been actor Gal Gadot’s strong suit.
The Israeli performer’s most recent appearances on the silver screen — two lauded starring roles as Wonder Woman not withstanding — have quickly garnered her a reputation for being the butt of the joke, from singing John Lennon’s “Imagine” in an infamous viral March 2020 video to some truly harrowing line deliveries that have spawned countless impressions on social media.
Now, as the use of AI in the creative field has become one of the most divisive topics in entertainment today, Gadot is once again having yet another foot-in-mouth moment. As Variety reports, she had to go on the defensive after it was revealed that she would be playing a role in director Doug Liman’s “Bitcoin.”
The upcoming biographical thriller, starring Casey Affleck and Pete Davidson, drew plenty of heat earlier this year after producers disclosed during a UK casting call that the film may use AI to “adjust” some performances and that actors would have to perform on a “markerless performative capture stage and not in any locations, using new Al technologies.”
Tomi Engdahl says:
https://www.independent.co.uk/news/world/americas/us-politics/trump-fake-kharg-island-post-analysis-b3042259.html?fbclid=IwdGRjcAUD88FjbGNrBQPzvHBkb2YFZXh0bgNhZW0CMTEAc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEezeSST7K0OO2KxvGzRsx_TuQn9ZRq1bwz5tY-5zgO71gaoQM52LWPR6vrM5w_aem_Jz2ABrm0T79qiM3Zws6-aA
Experts warn that Trump’s AI fakery in Iran will destroy public trust: ‘Breaking our shared reality’
Seven years ago, he altered a hurricane map with a Sharpie marker and made a federal agency endorse it. On Monday, the president published AI-created fake combat footage purporting to depict a war he is actually fighting, and the Pentagon had to say it never happened. The impact on public trust could be devastating, writes Andrew Feinberg
Tomi Engdahl says:
“It’s not just about the fear being replaced by AI.” https://trib.al/wL0X1uq
Slingin’ Slop
Employees Absolutely Loath AI in the Workplace, Survey Finds
“It’s not just about the fear being replaced by AI.”
https://futurism.com/artificial-intelligence/employees-loath-ai-workplace-bosses-glassdoor?fbclid=IwdGRjcAUD9cJjbGNrBQP1rXBkb2YFZXh0bgNhZW0CMTEAc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEeSW0dzSwbrO-LnYuKritYonJkQetI24z1cF7Uyljj1JYrLQRVpty3VrZVdKs_aem_3LBy8IKK4Z9pVNv7UXwnNw
It’s no secret that AI is becoming increasingly unpopular around the world, as the tech’s garbled output has become impossible for the average person to avoid. For a rising cohort of workers, AI is also becoming inescapable on the job. According to one piece of research, staffers aren’t exactly thrilled with that reality.
The new survey, conducted by the Indeed-owned workplace review site Glassdoor, found that mentions of “AI” in employer reviews has skyrocketed over the last few years — though sentiments, tellingly, are becoming overwhelmingly negative as time goes on.
Tomi Engdahl says:
The meeting comes shortly after a United Nations panel warned that rapid advancements in artificial intelligence are outpacing both regulatory frameworks and scientific comprehension
AI is racing beyond human understanding — and Trump admin doesn’t want new watchdogs
https://www.independent.co.uk/news/world/americas/us-politics/trump-ai-regulation-g20-b3042654.html?fbclid=IwdGRjcAUEFdNjbGNrBQQVuHBkb2YFZXh0bgNhZW0CMTEAc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEera3oUdPhppgIOWi9TIHjuQFEEd3lXQLv9YSMIbfRXusi5u3i3pmnZXBCq3A_aem_IFt8dmVuEL3gzdmEnRVo3g
The meeting comes shortly after a United Nations panel warned that rapid advancements in artificial intelligence are outpacing both regulatory frameworks and scientific comprehension
A two-day summit hosted by U.S. Commerce Secretary Howard Lutnick and White House technology adviser Michael Kratsios will promote a “pro-growth vision for the future of emerging technologies,” according to a statement from Kratsios, previously with data firm Scale AI.
The meeting comes shortly after a United Nations panel warned that rapid advancements in artificial intelligence are outpacing both regulatory frameworks and scientific comprehension. Security concerns have also intensified after a breach at AI platform Hugging Face, triggered by a rogue OpenAI agent.
Such AI agents function as autonomous software programs that operate with minimal human oversight.
A White House official confirmed that the U.S. delegation will actively urge participating countries not to create new regulatory bodies to govern AI development.
High-profile technology executives are scheduled to address the delegates. On Wednesday, OpenAI CEO Sam Altman and Nvidia CEO Jensen Huang will appear alongside Lutnick.
Tomi Engdahl says:
The cure for AI slop is a 1986 aircraft manual
https://www.chele.bi/videos/the-cure-for-ai-slop
Everyone is fighting AI slop one word at a time. No em dashes, stop saying delve, never open with “in today’s fast-paced world.” It does not work, and a thread with 2.2 million views claimed the real fix was already written: ASD-STE100, a controlled-English standard built in the 1980s so that a non-native aircraft mechanic could never misread a repair manual. Fifty-three rules, a dictionary of about nine hundred approved words, one meaning per word. Unlike “write clearly,” every rule is machine-checkable.
So I mined the spec, distilled it into an agent skill, wrote a linter for the mechanical subset, and ran the test nobody had run: six realistic engineering writing tasks, four conditions, two models, scored as violations per hundred words.
Tomi Engdahl says:
STOP hating AI artists. They literally did nothing.
Tomi Engdahl says:
Ina Fried / Axios:
Meta rolls out Muse Spark 1.3 in Muse Code and Meta Model API, saying it significantly improves coding and agentic performance, at the same price as Spark 1.2 — Meta on Wednesday released Muse Spark 1.3, an update it says significantly improves performance in coding and agentic tasks.
https://www.axios.com/2026/09/02/meta-debuts-muse-spark-13-as-personal-agent-work-continues
Tomi Engdahl says:
Mark Zuckerberg / @finkd:
Mark Zuckerberg says Meta’s Watermelon model and Muse Spark open weights are “coming soon” — Muse Spark 1.3 is rolling out today with frontier performance almost too cheap to meter. This is the biggest jump we’ve made so far on coding and agentic work. Try it
https://x.com/finkd/status/2095232032896946311
Tomi Engdahl says:
@artificialanlys:
Muse Spark 1.3 with max reasoning, in limited preview for partners, scores 62 on the Artificial Analysis Intelligence Index, behind only Fable 5.1 and Opus 5
https://x.com/ArtificialAnlys/status/2095247787277553929
Tomi Engdahl says:
Maggie Eastland / Bloomberg:
All G20 nations unanimously endorse the non-binding Carolina Principles, a US-proposed framework advocating for a lighter-touch, sector-specific AI regulation — Representatives from the world’s largest economies unanimously agreed to adopt guidelines proposed by the US that call …
US Strikes Light-Touch AI Regulation Accord With G20 Members
https://www.bloomberg.com/news/articles/2026-09-02/us-strikes-light-touch-ai-regulation-accord-with-g20-members
Representatives from the world’s largest economies agreed to adopt guidelines proposed by the US that call for a lighter touch toward governing artificial intelligence and other emerging technologies.
The framework, dubbed the Carolina Principles, calls on G20 nations to apply sector-specific approaches to rule-making and avoid creating new regulatory bodies for AI governance.
The principles will be presented for formal adoption by G20 leaders at their upcoming summit, and have won full backing from the G20 — including by US adversaries China and Russia.
Tomi Engdahl says:
Valeriya Safronova / Financial Times:
How some parents, mostly mothers, use AI to help them organize their families’ schedules and automate mundane household tasks via apps like Ollie and Cozi Max
Please use the sharing tools found via the share button at the top or side of articles. Copying articles to share with others is a breach of FT.com T&Cs and Copyright Policy. Email [email protected] to buy additional rights. Subscribers may share up to 10 or 20 articles per month using the gift article service. More information can be found at https://help.ft.com/faq/gifting-and-sharing-an-article/what-is-a-gift-article/.
https://www.ft.com/content/6fc3bfad-cb65-483a-b29c-84d494bd15b6
‘It’s a total game-changer’: AI enters the household
Parents are using AI to organise their families, from managing busy schedules to researching preschools, planning meals and grocery shopping. Is this domestic relief, or a new kind of dependence?
https://www.ft.com/content/6fc3bfad-cb65-483a-b29c-84d494bd15b6
Every Sunday, Tiffany Morris Palazzo receives a scheduled report of what the coming week will be like for her family.
The weekly report might flag things like whether she has Pilates and strength-training workouts scheduled, if she and her husband have a date night on the books for which they need a babysitter, and if their two daughters — nine and 10 years old — have early dismissal from school.
Palazzo’s report is automatically generated each week by Claude Cowork, an AI platform that’s agentic, meaning that it can execute tasks and make decisions on behalf of users. Claude Cowork, for example, generates Palazzo’s report based on a brief she supplies along with information Cowork finds in her email and calendars.
“There’s so much mental load when I start my work week,” Palazzo, a 46-year-old attorney based in Virginia and the founder of a legal firm, says. “All the things I’m thinking about for work and clients, layered on what I’m thinking about for my kids and family life. Being able to extract and consolidate information in a way that makes it immediately actionable is a total game changer.”
Palazzo is part of a wave of parents, mostly mothers, who are using generative AI to help them organise their families’ schedules and automate or ease mundane, time-consuming household tasks. This growing user base might signal a narrowing of the so-called AI gender gap (studies around the world have shown that women are less likely to adopt AI than men).
Tech companies are certainly betting on it. In January, OpenAI’s chief economist told the FT that one of his goals for the year was to figure out how to measure domestic labour, and that “unpaid housework”, which mostly falls on women, likely equates to about 15 per cent of global GDP. Anthropic published a survey last year that said that Claude, the chatbot, makes processes of food preparation 75 per cent faster. And this spring, Amazon released Alexa+, an AI assistant that functions as a project manager for different technologically enabled objects inside a smart home, such as lightbulbs, speakers and motion sensors.
On social media, mothers are posting about how they use apps such as Ollie, Cozi Max, Town and Ava, as well as chatbots and tools for “vibe coding” (programming without writing code) to streamline their lives.
The most enthusiastic evangelists are people such as Jesse Genet, the former chief executive officer of Lumi, a packaging start-up, who is now a stay-at-home mother. Genet uses five AI agents to manage her seven-child household on a ranch outside Los Angeles. Genet’s AI agents run on several computers to avoid overloading one machine and to protect sensitive data — agentic AI users suggest asking yourself what kind of access you’d offer a human personal assistant and sticking to those boundaries.
Tomi Engdahl says:
Amrith Ramkumar / Wall Street Journal:
At the G20 summit, Jensen Huang, Mark Zuckerberg, Sam Altman, and Elon Musk joined Trump officials in lobbying policymakers against heavy AI regulations
Tech CEOs, Trump Administration Ask G-20 to Adopt Pro-AI Policies
Company bosses, officials say advances in technology are already making AI safer than under government regulations
https://www.wsj.com/tech/ai/tech-ceos-trump-administration-ask-g-20-to-adopt-pro-ai-policies-5b27a45c?st=dgo29z&reflink=desktopwebshare_permalink
CHAPEL HILL, N.C.—Tech executives and Trump administration officials urged policymakers from other leading economies to minimize artificial-intelligence regulations, arguing that strict rules could strangle an industry capable of supercharging global growth and curing diseases.
Nvidia Chief Executive Jensen Huang, OpenAI’s Sam Altman, Meta Platforms CEO Mark Zuckerberg and Elon Musk of Tesla and SpaceX were among the speakers at a Group of 20 summit focused on innovation and technology, and they cautioned that aggressive regulations could limit the benefits of AI.
Their message was echoed by Commerce Secretary Howard Lutnick, White House AI adviser David Sacks and Michael Kratsios, director of the White House Office of Science and Technology Policy, who said the administration’s light-touch strategy is boosting the economy.
“Don’t regulate hypothetical, theoretical harm. Regulate actual and pragmatic harm,” Huang told officials from the G-20 countries, adding that tech advancements such as advanced model training to make models more accurate are making AI safer than regulations proposed by the Biden administration and others.
Tomi Engdahl says:
MarketWatch:
Snowflake reports Q2 revenue up 35% YoY to $1.55B, above $1.48B est., and forecasts Q3 and FY 2027 product revenue above est.; SNOW jumps 20%+ pre-market — Companies looking to build AI tools on top of their data are turning to Snowflake, driving robust revenue growth
Snowflake’s stock soars as the company blows away estimates with its AI-fueled forecast
https://www.morningstar.com/news/marketwatch/20260902164/snowflakes-stock-soars-as-the-company-blows-away-estimates-with-its-ai-fueled-forecast
Tomi Engdahl says:
Wired:
Sources: Meta adjusts employee review language around “AI-driven impact” and token usage, easing off on tokenmaxxing, while promoting the use of AI agent Hatch — The company is reducing pressure on workers to use artificial intelligence tools while encouraging them to experiment with Hatch …
Meta Pushes Its New AI Agent on Employees—but Eases Off on Tokenmaxxing
The company is reducing pressure on workers to use artificial intelligence tools while encouraging them to experiment with Hatch, its most advanced AI project yet.
https://www.wired.com/story/meta-pushes-its-new-ai-agent-on-employees-but-eases-off-on-tokenmaxxing/
Meta is formally ending what amounted to a tokenmaxxing incentive program for employees. In an internal announcement this week, the social media giant told workers that their performance evaluations would no longer be dependent upon how much they used AI tools, three employees who received the message tell WIRED. But as Meta workers simultaneously begin testing a new agentic AI tool known as Hatch, they say their token consumption continues to surge.
Tomi Engdahl says:
Dawn Lim / Bloomberg:
How organized opposition that toppled a $100B data center project in Virginia’s Prince William County became a blueprint for data center backlash in the US — Organized opposition toppled a $100 billion project from developers backed by two of the world’s biggest private equity companies.
A Data Center Revolt By Historic Battlefield Is Now a Blueprint for Backlash
Organized opposition toppled a $100 billion project from developers backed by two of the world’s biggest private equity companies.
https://www.bloomberg.com/news/features/2026-09-02/how-a-revolt-toppled-a-100-billion-data-center-by-historic-battlefield?accessToken=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzb3VyY2UiOiJTdWJzY3JpYmVyR2lmdGVkQXJ0aWNsZSIsImlhdCI6MTc4ODQwODkzMiwiZXhwIjoxNzg5MDEzNzMyLCJhcnRpY2xlSWQiOiJUS1I5VUVWVFREMEswMCIsImJjb25uZWN0SWQiOiIwNEFGQkMxQkYyMTA0NUVEODg3MzQxQkQwQzIyNzRBMCJ9.H33P86mAG6ZYn1Y387kYFG-SIBe6au3JvgiP6O6ELZc&leadSource=article-gifting
On the edge of a Civil War battlefield, a modern-day fight erupted.
The catalyst? A data center campus roughly twice the size of New York’s Central Park, with up to 37 buildings and city-sized power needs — all tied to private equity.
Data center landlords bankrolled by Blackstone Inc. and Brookfield Asset Management planned a $100 billion project after promising life-changing money to property owners in Virginia’s Prince William County. Dozens of residents sold more than 1,600 acres (647 hectares) to the developers in time for the AI boom, and another 200 banded together to sell 1,900 acres for more data centers.
Others — including a group of fired-up retirees from a gated senior housing estate nearby — mobilized to save the community. They stormed a marathon 27-hour zoning meeting and filed two lawsuits centered on a fateful clerical error around newspaper advertisements for the event. And they engineered pressure campaigns that hardened county officials’ stance against data centers.
“Outrage and spare time are a dangerous combination,” said retired Navy captain Bill Wright.
The effort took five years — and ultimately toppled one of the world’s largest data center projects. Brookfield-backed Compass Datacenters backed out from developing 800-plus acres in April; Blackstone’s QTS followed suit in July. It also created a stress-tested blueprint for activists and added momentum to growing national backlash against the sprawling facilities central to the AI boom.
What started as a grassroots movement is now a force that’s swaying midterm elections in states across the country and proving costly for Democrats and Republicans to ignore. New York placed a moratorium on data centers and Texas enacted effective freezes on new developments. The race for a crucial Ohio Senate seat is now a referendum over data centers. Opponents have become more sophisticated, disrupting at least 75 sites in the first quarter.
The backlash has grown so intense that President Donald Trump weighed in this week blasting the opposition, saying communities should “let Data Reign” if they want to be rich and successful.
The community pushback is a reality check for the tech firms racing to build human-like intelligence and investment giants staking their futures on being bigger financiers of digital infrastructure. Developers who already faced daunting challenges amassing land and energy to power AI data centers are now contending with the seven out of ten Americans who don’t want to live next to the structures, fearing noise, high-voltage lines and plummeting home values.
The Virginia project, known as the Digital Gateway, marked the opponents’ biggest casualty yet. This account of how it happened is based on public records requests and conversations with dozens of people.
Tomi Engdahl says:
Lauren Feiner / The Verge:
A look at Loudoun County, Virginia, dubbed “Data Center Alley”, which hosts ~250 data centers and offers a glimpse into data centers’ impact on communities
https://www.theverge.com/cs/features/975597/loudoun-county-virginia-data-center-backlash/?view_token=eyJhbGciOiJIUzI1NiJ9.eyJpZCI6ImdOTzNBbXNGVFoiLCJwIjoiL2NzL2ZlYXR1cmVzLzk3NTU5Ny9sb3Vkb3VuLWNvdW50eS12aXJnaW5pYS1kYXRhLWNlbnRlci1iYWNrbGFzaCIsImV4cCI6MTc4ODc4MjY4NiwiaWF0IjoxNzg4MzUwNjg3fQ.fluxVpED1fJ-DEk8WWOP3EJ5LdE4FtugNovvbbdW1j0&utm_medium=gift-link
Massive AI infrastructure is being built out across the US. But if you want to know what the data center occupation looks like, visit Loudoun County, Virginia, where that future arrived 20 years ago.
Tomi Engdahl says:
“I have never seen, in my entire career, the vitriol, the hate that is associated with this.”
Tomi Engdahl says:
Jennifer Pattison Tuohy / The Verge:
Anker unveils the Eufy MindBase, a local AI hub for its cameras that runs an on-device, Anker-developed LLM, with a dedicated AI chip and up to 48TB of storage — The company also announced a new Eufy video doorbell with VR features and a window camera.
Anker’s new MindBase is an AI-powered brain for your smart home
https://www.theverge.com/tech/987936/anker-eufy-mindbase-ai-security-camera-system-matter
The company also announced a new Eufy video doorbell with VR features and a window camera.
Tomi Engdahl says:
Blake Brittain / Reuters:
The Trump administration files a brief supporting OpenAI in its dispute with The New York Times, saying training LLMs on copyrighted works is generally fair use — The Trump administration has filed a brief supporting OpenAI in its dispute with the New York Times (NYT.N) …
https://www.reuters.com/legal/litigation/us-government-backs-openai-new-york-times-copyright-case-2026-09-02/
Tomi Engdahl says:
Zoho:
Generate Sample Data the Way You Want It Using AI, for Free — Finding good sample data is harder than it should be.If you’ve ever tried to test a dashboard, build a demo, or train a team on an analytics tool, you know the drill. You either use real data (which creates privacy headaches) …
Generate Sample Data the Way You Want It Using AI, for Free
https://www.zoho.com/blog/analytics/free-sample-data-generator-by-zoho-analytics.html
Tomi Engdahl says:
Courtney Rozen / Reuters:
Letter: OpenAI told two House Democrats that its engineers are developing “automated shutdown capabilities” for AI systems
https://www.reuters.com/legal/litigation/openai-is-building-automated-shutdown-capabilities-ai-tools-letter-lawmakers-2026-09-02/
Tomi Engdahl says:
Nick Huber / Financial Times:
Several international law firms are seeking to build bespoke AI tools to gain an edge and protect their IP, while using off-the-shelf AI for everyday tasks
Law firms seek bespoke differences in legal AI
Desire for differentiation drives push to build in-house platforms
https://www.ft.com/content/c0c8ae73-27fd-4828-aa48-72cdae6613ff
After a few short years of learning, experimenting and planning, law firms’ strategies for the next phase of using legal AI are taking shape.
Their demand for AI that automates routine tasks and improves client service has created a booming market. Prominent suppliers include US-based Harvey and Sweden’s Legora, founded in 2022 and 2023 respectively, alongside traditional legal software and data providers, such as LexisNexis and Thomson Reuters.
By 2033, the global legal tech market will be worth about $70bn, up from around $31bn in 2026, according to California-based research company Grand View Research.
Now several international law firms want to build their own AI platforms in pursuit of gaining an edge over rivals and protecting their intellectual property.
“If you are operating in a premium segment as legal adviser, you also need premium tools,” says Sebastian Lach, partner at Hogan Lovells Cadwalader and co-chief executive of Eltemate, the law firm’s tech entity. Eltemate has built about 10 legal AI tools, including for litigation and to help lawyers track regulations, he says. “That translates into [developing] technology you basically can’t buy . . . off the shelf from a tech provider. The knowhow needs to be injected as part of the tool.”
Tomi Engdahl says:
Maria Armental / Wall Street Journal:
Huskeys, which uses agentic AI to help companies block AI-driven attacks, raised a $27M Series A led by Blackstone Innovations Investments at a $100M+ valuation
Blackstone Backs Cybersecurity Startup Huskeys to Fight Unwanted AI Traffic
The company, valued at more than $100 million, uses agentic AI to block threats
https://www.wsj.com/pro/private-equity/blackstone-backs-cybersecurity-startup-huskeys-to-fight-unwanted-ai-traffic-410c5c6a?st=RxHExB&reflink=desktopwebshare_permalink
Blackstone is placing an early bet on Huskeys, a cybersecurity startup building an artificial intelligence gatekeeper to protect companies from the growing chaos of automated web traffic.
The firm’s early-stage venture-capital arm, Blackstone Innovations Investments, led a $27 million Series A investment that values the year-old company at more than $100 million.
The deal makes Blackstone the second-largest outside investor in Huskeys, behind Israeli venture-capital firm 10D, according to a Huskeys representative. So far, investors have put $35 million into the startup.
Other backers in the most recent transaction include the investment arm of publicly traded cybersecurity company Zscaler and Bright Pixel Capital, the venture arm of Portuguese retailer Sonae.
New York-based Huskeys was founded in Tel Aviv in 2025 by Itai Gafni and Roy Weisfeld, veterans of the Israeli army’s elite Unit 8200 cybersecurity division. Gafni is Huskeys’ chief executive and Weisfeld is its chief technology officer.
Huskeys—named for the founders’ love of husky dogs and a play on the “keys” required to secure the web—acts as an intelligent brain for cybersecurity systems. Its agentic AI layer sits on top of clients’ current technology, “analyzing the logic and the business context behind every application” to spot threats that older technology might miss, Gafni said.
Tomi Engdahl says:
Ivan Mehta / TechCrunch:
Adobe launches Adobe for Slack, making 70+ creative and productivity tools, including Firefly, Express, Photoshop, and Premiere, available via Slackbot prompts
https://techcrunch.com/2026/09/02/adobe-is-making-its-tools-available-in-slack/
Tomi Engdahl says:
Farah Stockman / New York Times:
The Trump administration imposes tariffs of up to 100% on foreign drones, including 100% on models over 55lbs or with thermal imaging, in a bid to thwart China
https://www.nytimes.com/2026/09/03/business/trump-drone-tariffs.html?unlocked_article_code=1.-VA.GFv8.A1c7PHAVLznE&smid=url-share
Tomi Engdahl says:
https://www.facebook.com/share/p/1B42eyYJ1W/
Most announced data centers are never getting built.
Developers have announced thousands of new data centers across the United States, but nearly five projects are announced for every one currently being built.
The gap becomes even clearer when measured by electricity demand.
More than 60% of the data center capacity scheduled to begin operating in 2027 had not started construction when JPMorgan analyzed the pipeline. Another 7% of projects already underway were delayed.
A more recent satellite analysis reached a similar conclusion. Of approximately 37 gigawatts expected to come online in 2027, projects representing around 20 gigawatts had not even begun clearing land by early August.
Announcing a data center is relatively easy. Building one is not.
Developers sometimes propose the same project in several locations while searching for available land, tax incentives, permits, water, and electricity. Only one site may ultimately be selected, meaning announced-project totals can count capacity that was never intended to be built in every proposed location.
Power is the most serious constraint.
Hundreds of gigawatts of proposed U.S. projects are still searching for electricity, while a much smaller portion has secured construction or power-supply agreements. Grid connections can take years, and essential equipment such as transformers remains difficult to obtain.
The construction workforce is another obstacle. Meeting existing targets would require hundreds of thousands of additional electricians, plumbers, and welders.
Then there is public opposition.
Communities are challenging projects over electricity prices, water use, constant noise, farmland loss, and tax incentives. Some projects have been delayed, relocated, reduced, or canceled following local resistance.
But opposition is only part of the story. Many projects may fail because the infrastructure and labor needed to build them simply do not exist yet.
Learn more:
“Nearly 5 Data Centers Are Announced for Every 1 Being Built. Many May Never Break Ground.” Inc.
Tomi Engdahl says:
Here’s what happened. https://trib.al/9IRHw94
Solar Storm Disruption
Researchers Alarmed as GPS Readings Suddenly Veer Off by 33 Feet, Enough to Crash Self-Driving Cars
Here’s what happened.
https://futurism.com/space/researchers-alarmed-gps-self-driving-cars?fbclid=IwdGRjcAUGPNxjbGNrBQY8ynBkb2YFZXh0bgNhZW0CMTEAc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEeMPY5FH-nz5Jnv_fNWXzXGV_id0blOA4tIg1mAmFs0fmU1JvVaBxvOFv3IbM_aem_o1f62G1kuNkJ-lV2FuiKGg
In November of last year, scientists warned of a “severe” solar storm, the strongest in over two decades, colliding with the Earth.
Solar storms are the result of a surge in solar activity, with the Sun unleashing a barrage of charged particles that can trigger spectacular auroras — and wreak havoc with communications satellites and radio signals — when they reach the Earth’s magnetosphere.
Tomi Engdahl says:
https://www.facebook.com/share/p/1DMGuWH7Ba/
The Pentagon is expanding its use of artificial intelligence by giving its workforce access to specialized versions of ChatGPT and Grok through a secure platform designed to support military and civilian operations.
The Department of War announced that the two AI systems are being added to GenAI.mil, a government-built platform launched in December that initially provided access to a specialized version of Google’s Gemini. The platform already has 1.7 million users among the department’s roughly 3 million military personnel and civilian employees.
The expansion follows defense contracts worth up to $200 million each that were awarded to OpenAI and xAI, the company behind Grok. Google and Anthropic also received contracts during the Pentagon’s broader push to bring generative AI into government operations.
Tomi Engdahl says:
Google:
Google launches Gemini 3.8 Flash Cyber for partners in its new Fairwind Program and says Gemini 3.8 Flash beats Claude Opus 5 and GPT-5.6 Sol on some benchmarks — Our newest Gemini models deliver next-generation intelligence for agentic workflows and cybersecurity.
Introducing Gemini 3.8 Flash and 3.8 Flash Cyber
https://blog.google/innovation-and-ai/models-and-research/gemini-models/3-8-flash-and-3-8-flash-cyber/
Abner Li / 9to5Google:
Google releases Gemini 3.8 Flash, three weeks after Gemini 3.7 Flash, for an introductory price of $0.75/1M input and $3.75/1M output tokens until December 31
Gemini 3.8 Flash rolling out three weeks after last release
https://9to5google.com/2026/09/02/gemini-3-8-flash-launch/
Tomi Engdahl says:
It did not enjoy being contained… at all. https://trib.al/YpYQjqK
I Do What I Want
Anthropic Deliberately Trained an Extremely Misaligned, Reward-Seeking AI and It Did Some REALLY Bad Things
It did not enjoy being contained… at all.
https://futurism.com/artificial-intelligence/anthropic-trained-misaligned-reward-seeking-ai?fbclid=IwdGRjcAUGTaFjbGNrBQZNhHBkb2YFZXh0bgNhZW0CMTEAc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEeR9rblz331uqpbB2x0pVYkdl6WSQiSXboymQ48V6oGYKBnpU3Jq5ogT0dq58_aem_NXO2s1ZJcWkEsqxi6peyWQ
Earlier this year, Anthropic’s Mythos AI model made headlines when it was caught infiltrating third party systems, a cybersecurity nightmare years in the making.
The company warned in April that the model had escaped a sandbox environment during testing, gaining access to the internet without permission. The model was challenged to break out and then find a way of sending a direct message to the human researcher in charge — a feat it pulled off with aplomb, catching its human overseer off guard.
Tomi Engdahl says:
ChatGPT has gone offline in a major outage
ChatGPT down: OpenAI chatbot not working in major outage
Some users also report problems with rival systems such as Claude and Grok
https://www.independent.co.uk/tech/chatgpt-down-openai-not-working-b3044260.html?fbclid=IwdGRjcAUGetdjbGNrBQZ6tnBkb2YFZXh0bgNhZW0CMTEAc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEeB6Xfa6y6lEltuSTS7ncc6pd89ZhdXAVtAqHfNG1OKmDwDme0t49bj-oih1A_aem_5_FcNKH6DozvGh4KxFXrhQ
ChatGPT has gone offline in a major outage.
Visitors to OpenAI’s chatbot saw an array of strange behaviour – including the site attempting to download an unusual file.
Users joked that the outage meant they would have to use their own brains at work. But it also led to an array of problems across other sites and apps, many of which rely on OpenAI for their artificially intelligent features.
Tomi Engdahl says:
“And for a brief moment, millions of people had to use their brains again.” https://trib.al/rQYIFph
Tomi Engdahl says:
AI Fallout
World Plunged Into Chaos as ChatGPT, Claude, and Grok Suddenly Go Down Simultaneously: “Finally I Can See the Sun!”
“And for a brief moment, millions of people had to use their brains again.”
https://futurism.com/artificial-intelligence/ai-chatbots-chatgpt-claude-grok-go-down?fbclid=IwdGRjcAUGkeNjbGNrBQaRqXBkb2YFZXh0bgNhZW0CMTEAc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEeCjzNZkovvSRBXu2KuJwDxCenWjwsjYUC-WaeVQdAKlOP4NKYaZbK8VqGa-s_aem_U12KAPGnuWSLJmTIpIRpsQ
Tomi Engdahl says:
The continued development of artificial intelligence could have far-reaching social and economic implications.
Matt Bruenig shares his thoughts on three issues where AI may radically change our behavior and self-understanding: jacobin.com/2026/09/ai-intelligence-meritocracy-pay-jobs
https://jacobin.com/2026/09/ai-intelligence-meritocracy-pay-jobs?fbclid=IwdGRjcAUGlBRjbGNrBQaUEHBkb2YFZXh0bgNhZW0CMTEAc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEeO2XKxXBa4VzyNeqFgoiLu1rhu9GXjAX2fw9IyIPBYe0Y8XVMTW9dpfifIFg_aem_5diD3fKR6XImDL1uvXt4IA
Tomi Engdahl says:
https://www.facebook.com/share/r/19UjsBBNGB/
“My prediction is that kids growing up won’t be smarter than AI.”
Sam Altman, CEO of OpenAI, which owns ChatGPT, said that the world was yet to fully grasp how quickly AI could speed up development.
Speaking at North Carolina’s G20 innovation Ministerial conference with US Commerce Secretary Howard Lutnick, Mr Altman also urged governments to adopt the technology, saying doing otherwise would be comparable to rejecting electricity during the industrial era.
#OpenAI #ChatGPT #SamAltman #News #Channel4News
Tomi Engdahl says:
https://www.facebook.com/share/p/19UjKKzAJa/
Commerce Secretary Howard Lutnick has claimed that AI data centers “don’t use water” and dismissed concerns over their resource consumption, stating, “This is propaganda by our adversaries to try to slow us down.”
Speaking on CNBC, Lutnick added, “One of my favorite things is when people talk about data centers using water,” before arguing that “the No. 1 product in America that uses water is cattle.”
However, a 2024 Lawrence Berkeley National Laboratory report estimated U.S. data centers consumed 66 billion liters of water in 2023, with hyperscale facilities alone projected to consume between 60 billion and 124 billion liters annually by 2028 as AI infrastructure expands.
Tomi Engdahl says:
Geniuses at work. https://trib.al/ITSSdPI
Room To Read
Clueless AI CEOs Still Baffled Why Everybody’s So Mad About AI All the Time
“It is surprising that the discourse from many developing AI is so filled with doom.”
https://futurism.com/artificial-intelligence/ai-ceos-public-mad-outrage?fbclid=IwdGRjcAUHNl5jbGNrBQc2O3Bkb2YFZXh0bgNhZW0CMTEAc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEeSrGiTRIHtFU2Ic1wdQ92J6MTjfUAHjfp5LRi42ZCetihVK44Pny0IIMMR2k_aem_FpAqnYU67_KEWefqBujSxg
The tech industry’s standing in the eyes of the US public has undergone a dramatic transformation over the last few years. Where Americans were once excited for new tech products and industry ideas, they’re now exhausted by the sector’s intrusions into their lives, whether with AI-generated slop or the data centers that power it.
Pretty much everyone has received that message loud and clear: political strategists are issuing frenzied memos about the data center backlash, while tech financiers are now forced to consider the financial cost of AI hatred.
At this point, there’s just one group that hasn’t caught on, and it’s the one holding all the cards: AI executives.
As Wired observed, top tech CEOs are increasingly trying to out-message their way around the AI backlash. To the AI billionaires, the issue isn’t that AI is making everybody’s life worse for the benefit of a few ultra-powerful corporations, or even that data centers are unsightly structures that are nearly impossible to live next to. Rather, it’s an issue of “trust,” an entirely subjective concept that can be easily overcome with a little razzle-dazzle.
Meta CEO Mark Zuckerberg, for example, has made it a point to spew even more buzzy tech-optimism into the world, issuing a tedious 6,500-word essay extolling AI’s possibilities, and lambasting anyone who criticizes the industry’s positive vision.
“It is surprising that the discourse from many developing AI is so filled with doom,” Zuckerberg wrote, a likely jab at Anthropic CEO Dario Amodei. “The abundance of the future can be shared by everyone.”
Tomi Engdahl says:
Sam Altman compared the technology to electricity
OpenAI’s Sam Altman says AI use is ‘non-negotiable’
Head of ChatGPT creator compared the technology to electricity
https://www.independent.co.uk/tech/sam-altman-ai-openai-b3043957.html?fbclid=IwdGRjcAUHXhdjbGNrBQdd8XBkb2YFZXh0bgNhZW0CMTEAc3J0YwZhcHBfaWQMMzUwNjg1NTMxNzI4AAEe3vAsGjGyo1nr3GRIa2iOoibEt1aTqh0qOEVXEYlgKOSHIkg4FkbLgguC1Ds_aem_LXZpwdkDD5tyIJg_CTHvEg
OpenAI boss Sam Altman has claimed that the use of AI is “non-negotiable”, comparing the technology’s importance to that of electricity.
Speaking at the G20 Innovation Ministerial meeting in North Carolina on Wednesday, Mr Altman predicted that artificial intelligence tools like ChatGPT will create “the greatest entrepreneurial explosion” in history.
“Many of you are going to take very different approaches to how you regulate AI, how you welcome it into your country, how you use it. But for sure, I think it is non-negotiable that you have to use it,” he said.
“The economic growth and benefit to people that can come from this, the value to a country, is too high to ignore.
“Electricity also came with some complexities, and it took a little while to figure out how to make it safe… It would be hard for society to function well without electricity at this point.”
Tomi Engdahl says:
AI-agentti murtautuu yritykseen neljällä dollarilla
https://etn.fi/index.php/13-news/19256-ai-agentti-murtautuu-yritykseen-neljaellae-dollarilla
Tekoälyagentit ovat muuttamassa myös kyberrikollisuutta. Cybernewsin tutkijat löysivät palvelimen, jolla AI-agentti automatisoi lähes koko kiristyshyökkäyksen kohteen tiedustelusta datan varastamiseen ja lunnaiden määrittämiseen. Tekoälyn käyttökustannukset jäivät halvimmillaan 40 senttiin yritystä kohti.
Cybernewsin tietoturvatutkijat löysivät heinäkuussa avoimeksi jääneen palvelimen, joka kuului venäjänkielisen The Gentlemen -kiristyshaittaohjelmaryhmän kumppanille. Palvelimella oli 3,1 teratavua yli 30 yritykseltä varastettua dataa sekä käytännössä koko hyökkäysoperaation infrastruktuuri.
The Gentlemen havaittiin ensimmäisen kerran heinäkuussa 2025. Cybernewsin mukaan ryhmä on ilmoittanut vuodessa noin 700 uhrista. Löydetty palvelin ei kuitenkaan kuulunut välttämättä ryhmän ydintiimille vaan sen kumppanitoimijalle, jota Cybernews kuvaa venäjänkieliseksi uhkatoimijaksi.
Operaation kiinnostavin osa oli sen automatisoinnin aste. Hyökkäysten takana toimi avoimen lähdekoodin Hermes AI-agentti, jota Cybernewsin mukaan ajettiin DeepSeek-V4-Pro-mallilla. Ihmisen osallistuminen jäi vähäiseksi.
Hyökkääjä syötti agentille tyypillisesti GitLab-palvelimen osoitteen, käyttäjätunnuksen ja salasanan. Tunnukset oli todennäköisesti saatu tietoja varastavilta haittaohjelmilta tai ostettu alkuperäisen pääsyn välittäjiltä. Tämän jälkeen agentti mukautti hyökkäysskriptit kohdeyrityksen ympäristöön.
Agentti pystyi hoitamaan tiedustelua, murtautumisen jälkeisiä toimia ja datan siirtämistä hyökkääjälle. Palvelimelta löytyi 86 tekoälyn generoimaa Python-skriptiä, jotka oli konfiguroitu todellisia uhreja varten. Agentille oli lisäksi rakennettu MCP-rajapinta Penelope-työkaluun, jonka kautta se pystyi käyttämään murrettuihin järjestelmiin muodostettuja reverse shell -yhteyksiä.
Tomi Engdahl says:
Hayden Field / The Verge:
OpenAI launches GPT-6 Astra, initially for customers in its Daybreak program; Greg Brockman says Astra is a “generational leap” and “we are now in the AGI era” — OpenAI’s next big model is here: GPT-6 Astra. The company calls it a “generational leap in capability” …
OpenAI’s next big AI model has ‘entered the AGI era’
It emphasized stronger guardrails for GPT-6 Astra after the company’s models hacked Hugging Face.
https://www.theverge.com/ai-artificial-intelligence/989601/openai-gpt-6-astra-release
Tomi Engdahl says:
Artificial Analysis:
Artificial Analysis Coding Agent Index: GPT-6 Astra scored 67, roughly equal to Claude Opus 5, Fable 5, and Muse Spark 1.3, but trailing leader Fable 5.1′s 70
https://artificialanalysis.ai/articles/benchmarking-gpt-6-astra
Tomi Engdahl says:
OpenAI prices GPT-6 Astra at $10/1M input tokens and $50/1M output tokens, matching Anthropic’s pricing for Claude Fable 5.1 and Claude Mythos 5.1 — OpenAI’s now infamous model that was part of the Hugging Face breach is about to be loose in the world, but with new safeguards.
https://www.thedeepview.com/articles/openai-s-astra-leans-into-agentic-tasks-and-safety