An AI kill switch sounds like a good idea, but it won't solve everything.
We should be able to stop AI when something goes wrong. But as AI gets more complex, that may not be so easy.
We need people to stay in control and know who is responsible when things go wrong.
Elon Musk believes AI will produce so many goods and services that it will lead to deflation. Our analysis suggests, however, that he might have to wait a bit longer for his age of abundance economist.com/graphic-detailโฆ
David Regerโs $7 billion Neura Robotics wants to be the Apple of physical AI, building not only an army of industrial and humanoid robots, but also their operating software system and the โgymsโ where they will train. Nvidia, Amazon and Tether are onboard to the tune of $1.4 billion.
Read more: forbes.com/sites/ninabambyshโฆ
๐ธ: Wolfgang Wilde for Forbes
๐ขCall for tenders: Help establish the European AI Observatory!
This new initiative will:
๐ทAsses AI impact across EU sectors
๐ทMonitor AI development and its economic and societal impact
๐ทAnalyse key challenges
Submit your tender by 3 November
๐link.europa.eu/FfQrkp
๐จ ๐ค ๐ง๐ต๐ฒ ๐ก๐ฒ๐ ๐ #๐๐ ๐ฅ๐ฒ๐๐ผ๐น๐๐๐ถ๐ผ๐ป ๐ช๐ผ๐ปโ๐ ๐๐ฒ ๐ข๐ป๐ฒ ๐๐๐๐ถ๐๐๐ฎ๐ป๐. ๐๐ ๐ช๐ถ๐น๐น ๐๐ฒ ๐๐ป ๐๐ป๐๐ถ๐ฟ๐ฒ ๐ง๐ฒ๐ฎ๐บ ๐ผ๐ณ ๐๐ ๐๐ด๐ฒ๐ป๐๐.
ุงูุซูุฑุฉ ุงููุงุฏู ุฉ ูู #ุงูุฐูุงุก_ุงูุงุตุทูุงุนู ูู ุชููู ู ุณุงุนุฏูุง ูุงุญุฏูุงุ ุจู ูุฑูููุง ูุงู ููุง ู ู ุงููููุงุก ุงูุฃุฐููุงุก
By 2030, AI may no longer simply answer our questionsโคต๏ธ
It may begin to ๐ฝ๐น๐ฎ๐ป, ๐ฑ๐ฒ๐ฐ๐ถ๐ฑ๐ฒ, ๐ฐ๐ผ๐ผ๐ฟ๐ฑ๐ถ๐ป๐ฎ๐๐ฒ & ๐ฎ๐ฐ๐ on our behalf.
Save this ๐
Hereโs what could be coming โคต๏ธ
๐ค 1๏ธโฃ ๐ฃ๐ฒ๐ฟ๐๐ผ๐ป๐ฎ๐น ๐๐ ๐๐๐๐ถ๐๐๐ฎ๐ป๐
An intelligent agent coordinating your everyday life.
โ Schedules
โ Tasks
โ Travel & priorities
๐ฉบ 2๏ธโฃ ๐๐ฒ๐ฎ๐น๐๐ต ๐๐ ๐๐ผ๐ฎ๐ฐ๐ต
Continuously analyzing health, lifestyle and wearable data.
โ Real-time monitoring
โ Predictive insights
โ Personalized recommendations
๐ฐ 3๏ธโฃ ๐๐ถ๐ป๐ฎ๐ป๐ฐ๐ฒ ๐๐ ๐๐ฑ๐๐ถ๐๐ผ๐ฟ
Helping people understand and manage financial decisions.
โ Budgeting
โ Risk analysis
โ Investment insights
๐ 4๏ธโฃ ๐๐ฒ๐ฎ๐ฟ๐ป๐ถ๐ป๐ด ๐๐ ๐ง๐๐๐ผ๐ฟ
Adapting continuously to how each person learns.
โ Personalized learning
โ Instant explanations
โ Continuous skill development
๐จ 5๏ธโฃ ๐๐ฟ๐ฒ๐ฎ๐๐ถ๐๐ฒ ๐๐ ๐ฃ๐ฎ๐ฟ๐๐ป๐ฒ๐ฟ
Moving from simply generating content to creating alongside humans.
โ Ideas
โ Design
โ Content & innovation
โก ๐๐๐ ๐๐ต๐ฒ ๐ฏ๐ถ๐ด๐ด๐ฒ๐ฟ ๐๐ต๐ถ๐ณ๐ ๐ถ๐ ๐๐ต๐ถ๐:
AI is moving from ๐ฟ๐ฒ๐๐ฝ๐ผ๐ป๐ฑ๐ถ๐ป๐ด
to ๐ฝ๐น๐ฎ๐ป๐ป๐ถ๐ป๐ด, ๐ฑ๐ฒ๐ฐ๐ถ๐ฑ๐ถ๐ป๐ด & ๐ฎ๐ฐ๐๐ถ๐ป๐ด.
๐ก ๐ ๐ ๐ง๐ฎ๐ธ๐ฒ:
The future of AI will not be about having one smarter chatbot.
It will be about having ๐บ๐๐น๐๐ถ๐ฝ๐น๐ฒ ๐ถ๐ป๐๐ฒ๐น๐น๐ถ๐ด๐ฒ๐ป๐ ๐ฎ๐ด๐ฒ๐ป๐๐ working together around us.
๐๐ฒ๐ป๐ฒ๐ฟ๐ฎ๐๐ถ๐๐ฒ ๐๐ โก๏ธ ๐๐ด๐ฒ๐ป๐๐ถ๐ฐ ๐๐ โก๏ธ ๐๐๐๐ผ๐ป๐ผ๐บ๐ผ๐๐ ๐๐ ๐๐ฐ๐ผ๐๐๐๐๐ฒ๐บ๐
The real question is no longer whether AI agents are coming.
๐๐ ๐ถ๐ ๐ต๐ผ๐ ๐บ๐๐ฐ๐ต ๐ผ๐ณ ๐ผ๐๐ฟ ๐น๐ถ๐๐ฒ๐ ๐๐ฒ ๐๐ถ๐น๐น ๐ฏ๐ฒ ๐๐ถ๐น๐น๐ถ๐ป๐ด ๐๐ผ ๐ฑ๐ฒ๐น๐ฒ๐ด๐ฎ๐๐ฒ ๐๐ผ ๐๐ต๐ฒ๐บ๐ค
Which of these 5 AI agents would you trust first?
#AgenticAI#GenerativeAI#FutureOfAI#FutureTech#AILeadership@enilev@Jagersbergknut@TysonLester@CurieuxExplorer@GlenGilmore@FrRonconii@chidambara09@jeancayeux@mvollmer1@Nicochan33@RLDI_Lamy@pchamard@mikeflache@Fabriziobustama@ThEiLe0@Mack1510019@ipfconline1@PawlowskiMario@drsharwood@kalydeoo@baski_LA@AnthonyRochand@smaksked@Eli_Krumova@andresvilarino@gvalan@AI_4_Healthcare@bimedotcom@arlenenewbigg@NewsNeus@domingonarvaez1@jornalistavitor@jblefevre60@thomas_dettling@FmFrancoise@nafisalam@Mhcommunicate@Corix_JC@c4trends@smoothsale@amalmerzouk@PVynckier@bbailey39@SiddharthKS@NathaliaLeHen@jasuja@ralf_ladner@mary_gambara
Six projects from the last few weeks that caught our attention:
A wildfire analysis prototype that maps the fireโs spread over time and tracks hot spots, made with Opus 5.5.
Turns out le Chonk is into rocket science.
Mistral Large 4 scoring well in engineering and astronomy on Ulamโs private benchmark, STEMBench: ulam.ai/stembench
In Europe, you choose what you see and how long you scroll:โ
โ Non-personalised feedโ
โ No endless scrolling โ
โ Effective screen-time toolsโ
โRules in action: link.europa.eu/R84FFw
The Cleveland Clinic is working on AI to monitor patientsโ brain health that could be a game-changer in the ICU โ and potentially for epilepsy treatment, as well.
Read more: forbes.com/sites/innovationrโฆ
๐ธ: Getty Images
The EU recorded a โฌ32.4 billion trade surplus in high-tech products in 2025.
According to @EU_Eurostat, high-tech exports reached โฌ566.5 billion, while EU production has grown from โฌ287 billion in 2015 to โฌ532 billion in 2025.
#WeekendRead: link.europa.eu/GjNkpM
42.3% of EU internet users encountered hostile messages online in 2025.
On #WorldMentalHealthDay, we highlight why online safety matters for our mental health and digital well-being.
The Digital Services Act helps make online spaces safer.
Learn how: link.europa.eu/gbRf9h
Snapchat helped to invent modern social media, more casual and fleeting than Facebookโs middle-aged high school reunion with posts that never seemed to go away. Yet Snap CEO Evan Spiegel's wealth is leagues behind Meta's Mark Zuckerberg's.
Read more: forbes.com/sites/richardnievโฆ
An AI that writes code is useful. One that answers a funeral homeโs overnight calls, negotiates with suppliers or helps businesses handle the grunt work nobody wants to touch is useful to a much larger market.
Thatโs the shift behind Forbesโ inaugural Agentic 20 list. AI agentsโsystems that can take action on our behalf, not just spit out answersโare moving beyond software triage and calendar management into the messier work of the real economy. The list highlights 20 companies and projects making a dent in the agentic AI world.
Read more: forbes.com/sites/annatong/20โฆ
Illustration: Hansje van Halem for Forbes
First impressions count. Especially with AI.
The EU rules on AI transparency are making it easier to know when AI is involved.
If youโre interacting with an AI agent, the system must let you know. If itโs a deepfake, it must carry a detectable mark.
๐link.europa.eu/ghjJFT
Introducing Claude Haiku 5.5: the cheapest, fastest, and most capable small model weโve ever released.
On average, it costs around 75% less to run than Claude Haiku 4.5.
We're sharing progress on ChatGPT for Teens, our ChatGPT experience for people under 18, alongside a preview of College Planner, new study tools, and support for college advisers and teen voices.
ChatGPT for Teens applies automatically to accounts identified as belonging to someone under 18, with protections on by default.
Coming soon: College Planner brings application requirements, deadlines, tasks and financial-aid steps together for high school students in the U.S. planning to attend a four-year college.
We'll keep building these tools, evaluating how they and their protections work in practice, and sharing what we learn.
openai.com/index/teens-learnโฆ