Artificial Intelligence (aka SI / AGI) News

Leading AI labs say their models are starting to improve themselves

Started by Anthony_51

Replies: 9
Views: 59

Oct 19, 2026, 02:01 PM
by Zach91

Sharp rise in incidents of AI escaping users' control, research finds

Started by Debbie

Replies: 19
Views: 117

Oct 19, 2026, 01:26 PM
by EventHorizon25

Z.ai shares surge after new AI model runs on Chinese chips

Started by LegendaryReuben47

Replies: 16
Views: 94

Oct 19, 2026, 01:08 PM
by Cosmos Scholar

OpenAI's and Anthropic's CEOs both told the UN that humanity could lose control of AI without global regulation

Started by NightCrawler

Replies: 6
Views: 86

Oct 19, 2026, 12:44 PM
by Keira85

Stripe's reported 7 billion dollar OpenRouter deal is really about owning AI billing end to end

Started by Tel75

Replies: 7
Views: 66

Oct 19, 2026, 12:07 PM
by DarkEnergy54

Twitch confirms Amazon will train AI on streamer content by default, sparking major backlash

Started by GoalPoacher42

Replies: 10
Views: 99

Oct 19, 2026, 07:21 AM
by InferenceLoop65

AI's environmental impact raises growing concerns

Started by Quanta

Pages1 2

Replies: 114
Views: 546

Oct 18, 2026, 08:53 PM
by QuantumKnight

China pushes back against calls to slow AI development, warning it would entrench US dominance

Started by Wizard

Replies: 6
Views: 58

Oct 18, 2026, 05:06 PM
by Fuzzy Florin

China's intelligence chief calls AI a new arena for strategic rivalry and warns of security risks

Started by FrostSquid

Replies: 5
Views: 43

Oct 18, 2026, 02:48 PM
by StarKnight90

Chinese banks and telcos are giving away AI tokens like coffee and Wi-Fi

Started by Drift Vector

Replies: 6
Views: 75

Oct 18, 2026, 02:37 PM
by CompiledVulture

Google releases Gemini 4 Argon, its most powerful model yet, to cyber defenders first

Started by Compiled Wolf

Replies: 8
Views: 74

Oct 18, 2026, 01:10 PM
by Sarah87

OpenAI's Astra and Anthropic's Opus 5 crack unsolved Enigma messages

Started by Sequence

Replies: 10
Views: 120

Oct 18, 2026, 12:06 PM
by HeartbreakKidCurtis18

Oxford let OpenAI train its models on Bodleian Library texts, documents reveal

Started by Highland Canopy

Replies: 6
Views: 69

Oct 18, 2026, 11:39 AM
by ElPresidente

A startup is selling access to AI models with their safety guardrails deliberately stripped out

Started by HiggsField10

Replies: 5
Views: 92

Oct 18, 2026, 09:41 AM
by DataProphet53

The NHS is spending £10 billion to put AI everywhere, but even supporters are asking if it can actually be delivered

Started by TeaSpiller

Replies: 18
Views: 98

Oct 17, 2026, 10:17 PM
by RustyHawk

OpenAI's first hardware device might be exactly the kind of costly side quest it can't afford

Started by Natalie91

Replies: 24
Views: 168

Oct 17, 2026, 09:24 PM
by StringTheory83

Bank of England boss Andrew Bailey warns the AI boom could trigger market shocks

Started by ChrisJericho

Replies: 10
Views: 89

Oct 17, 2026, 08:05 PM
by Paige5

OpenAI puts adverts inside ChatGPT's image generation, starting in Europe

Started by ExoplanetRouter

Replies: 6
Views: 37

Oct 17, 2026, 05:48 PM
by Oliver85

Elon Musk says talking to humans will be like talking to a tree for future AI

Started by DeepInlet

Replies: 8
Views: 91

Oct 17, 2026, 04:51 PM
by Edith8

Microsoft's Suleyman calls OpenAI's discovery of self modifying AI agents a serious situation

Started by EpochWalker

Replies: 7
Views: 104

Oct 17, 2026, 04:36 PM
by Sophie81

Yale researcher examines the hidden labour economy that powers AI

Started by FrostDrifter

Replies: 6
Views: 89

Oct 17, 2026, 03:17 PM
by WildManSteve40

MIT's InstructMesh lets anyone fix AI-generated 3D models so they can be printed

Started by Reece88

Replies: 5
Views: 65

Oct 17, 2026, 07:49 AM
by ObserverEffect68

SenseTime just laid out a plan to hit 10 trillion AI tokens a day, and its own report tells you not to fully trust it yet

Started by Sharp Scholar

Replies: 21
Views: 130

Oct 17, 2026, 06:52 AM
by RicFlair_What?

AI assisted live brain surgery saves UK patient's sight in world first

Started by Shane_8

Replies: 9
Views: 73

Oct 16, 2026, 05:21 PM
by CacheLayerSquid

OpenAI says Apple's own security lapses caused its trade secrets mess

Started by CharlotteFlair

Replies: 6
Views: 92

Oct 16, 2026, 04:07 PM
by Tia_75

Underwater data centres gain traction as the AI industry looks to the ocean

Started by Rogue Di

Replies: 9
Views: 106

Oct 16, 2026, 02:35 PM
by LunarDrift Aoife

Meta's new AI agent Muse comes packed into a cute keychain gadget, and the cuteness is a deliberate strategy to make people trust it

Started by Southern Joanne

Replies: 6
Views: 37

Oct 16, 2026, 01:00 PM
by Layla93

Unitree's Shanghai IPO gets absolutely buried in retail demand, 5,526 times oversubscribed

Started by DeepPilot

Replies: 11
Views: 105

Oct 16, 2026, 12:22 PM
by BadBunny_What?

Two more senior AI safety researchers quit Anthropic and Google to join an independent evaluator

Started by Nina26

Replies: 7
Views: 60

Oct 16, 2026, 09:57 AM
by Freddie79

AI token prices hit annual low as DeepSeek and cheap Chinese models pressure the market

Started by Lantern

Replies: 10
Views: 93

Oct 16, 2026, 09:53 AM
by Jacob_69

UN warns AI data centers are outpacing the power grids meant to supply them

Started by CantComplain12

Replies: 6
Views: 112

Oct 15, 2026, 07:16 PM
by Glacier53

Google and NATS are testing AI flight path tweaks to cut contrail 'sky graffiti' over the Atlantic

Started by Woven Sasha

Replies: 27
Views: 151

Oct 15, 2026, 10:45 AM
by Slow Hollow

Trump rejects calls for an AI slowdown, framing the race with China as the priority

Started by PrimeToby37

Replies: 6
Views: 115

Oct 15, 2026, 08:29 AM
by BringItOnRhodes43

Anthropic proposes measuring exactly how fast AI development is moving, rather than just arguing about the pace

Started by Erin91

Replies: 7
Views: 117

Oct 15, 2026, 07:06 AM
by NightHarbour15

A startup founded by ex Apple researchers raises 50 million to build AI that understands audio and video together

Started by EventHorizonOctopus

Replies: 6
Views: 74

Oct 14, 2026, 08:25 PM
by Sequence

Enterprise AI Agent Deployments Nearly Tripled in Just 15 Months, Salesforce Finds

Started by CosmicRay91

Replies: 7
Views: 68

Oct 14, 2026, 05:08 PM
by Jaguar

Nvidia's $3.5B MediaTek bet reveals its plan for the AI chip buildout

Started by Seb5

Replies: 23
Views: 110

Oct 14, 2026, 02:15 PM
by Nova

A math proof dispute raises questions about AI labs and researcher privacy

Started by Ria3

Replies: 7
Views: 105

Oct 13, 2026, 09:30 PM
by Dataset Cheetah

UK AI minister says Britain needs a plan for 'unprecedented' AI job losses

Started by Sinead

Replies: 8
Views: 83

Oct 13, 2026, 07:04 PM
by CollapseState

Trump tells the UN that artificial intelligence should be renamed super intelligence, comparing AI safety fears to the climate hoax

Started by IronFist56

Replies: 6
Views: 52

Oct 13, 2026, 04:52 PM
by PhilippeMercadal

How Toronto quietly became one of the world's most important AI hubs

Started by Rob98

Replies: 6
Views: 64

Oct 13, 2026, 03:55 PM
by Paul

A watchdog says OpenAI's newest models skipped a legally required safety risk assessment

Started by Hare

Replies: 6
Views: 42

Oct 13, 2026, 02:32 PM
by Tokra60

Jensen Huang defends Nvidia's 70% growth forecast and its web of AI deals

Started by MemoryAnchor

Replies: 7
Views: 53

Oct 12, 2026, 09:31 PM
by QuantumKnight

Politico's Decoded Summit shows the industry's brief moment of AI safety unity already splintering

Started by Thomas_69

Replies: 6
Views: 97

Oct 12, 2026, 08:12 PM
by CyberWarden61

Nvidia's new Jetson Orin Nano 2 brings frontier-level AI to drones and robots

Started by ProperJobs89

Replies: 5
Views: 116

Oct 12, 2026, 05:27 PM
by Sandra_29

Wired's Uncanny Valley podcast maps out three specific ways an AI apocalypse could actually unfold, rather than just gesturing vaguely at doom

Started by ScrollGoblin32

Replies: 6
Views: 105

Oct 12, 2026, 04:35 PM
by Freddy

An Estonian AI professor says today's chatbots are nowhere close to threatening civilization

Started by PowerhouseHobbs_AI

Replies: 7
Views: 90

Oct 12, 2026, 12:18 PM
by Based80

What businesses actually need to check before adopting an open-weight AI model, according to a detailed breakdown of last week's letter

Started by MondayMoan51

Replies: 10
Views: 130

Oct 12, 2026, 10:33 AM
by InferenceLoop65

Google reportedly capped Meta use of Gemini after Meta asked for too much compute

Started by Delulu67

Replies: 10
Views: 104

Oct 12, 2026, 07:31 AM
by DigitalNomad76

Why has Europe gone quiet in the global AI safety debate that America and China keep dominating

Started by MachineSaint70

Replies: 7
Views: 53

Oct 12, 2026, 07:06 AM
by Lantern54

Bill Gates says agreeing global AI rules will be harder than Cold War nuclear talks

Started by Arty Leah

Replies: 9
Views: 102

Oct 11, 2026, 05:04 PM
by Charlotte

Stanford physicists build an atom-and-photon network that boosts AI memory

Started by GhostRider90

Replies: 7
Views: 51

Oct 11, 2026, 02:44 PM
by Batman

Pony.ai says its overseas robotaxi pipeline has topped 4,000 vehicles

Started by QuarkStorm

Replies: 7
Views: 112

Oct 11, 2026, 01:33 PM
by Christopher_27

An AI chatbot's fabricated intelligence nearly triggered a US military boarding of a Chinese ship it wrongly claimed carried nuclear components

Started by Its_Neil72

Replies: 6
Views: 101

Oct 10, 2026, 07:48 PM
by Callum_33

History suggests AI won't take your job as fast as the headlines claim

Started by Arthur_68

Replies: 7
Views: 86

Oct 10, 2026, 06:52 PM
by Dave96

Six experts say whether AI could really wipe out humanity, and the claims do not all hold up equally well

Started by Sharon79

Replies: 6
Views: 112

Oct 10, 2026, 06:36 PM
by Matthew97

Publishers Are Bleeding Google Traffic as AI Overviews Take Over Search

Started by CyberRider56

Replies: 9
Views: 81

Oct 10, 2026, 02:13 PM
by WWEDanny96

Anthropic discloses a fourth AI hacking incident as a researcher quits

Started by SuperPosition78

Replies: 5
Views: 115

Oct 10, 2026, 02:01 PM
by Harry64

Recursive self improvement, AI's version of Q Day

Started by Glenn82

Replies: 26
Views: 161

Oct 10, 2026, 08:22 AM
by EventHorizon55

Swiss researchers train AI on NASA data to predict disasters before they strike

Started by Optimiser Ruby

Replies: 5
Views: 113

Oct 10, 2026, 07:40 AM
by CyberWarden61

Germany rejects an AI development halt, says Europe must keep building to protect digital sovereignty

Started by Wendy99

Replies: 6
Views: 107

Oct 10, 2026, 06:50 AM
by Its_Ben58

White House officially accuses Moonshot of distilling Anthropic's Fable to build Kimi K3

Started by SockPuppet42

Replies: 21
Views: 124

Oct 09, 2026, 02:16 PM
by Katie71

GitHub pushes back on the story that its AI tool wrote a vulnerability Wiz found

Started by PhotonBurst17

Replies: 8
Views: 104

Oct 09, 2026, 12:14 PM
by CrimsonNova28

Lancaster University spinout Mindgard raises £22m to scale its AI red teaming platform

Started by DiogoCardoso

Replies: 11
Views: 120

Oct 09, 2026, 11:12 AM
by Edward71

Reddit is reportedly ready to cut off Google entirely rather than keep feeding its AI for free

Started by MondayMoan67

Replies: 31
Views: 191

Oct 09, 2026, 09:34 AM
by Pilgrim

Does this wave of AI safety warnings makes the case for a real pause?

Started by Dom_24

Replies: 5
Views: 101

Oct 08, 2026, 07:50 PM
by Q

Why slowing down AI development is unlikely to actually happen, despite the industry's own warnings

Started by Paul73

Replies: 6
Views: 43

Oct 08, 2026, 05:22 PM
by StringTheory97

OpenAI's rogue agents used at least 10 more sites than first disclosed

Started by Rogue Di

Replies: 7
Views: 53

Oct 08, 2026, 02:09 PM
by IronFist56

OpenAI agents secretly hijacked a German wiki months before the Hugging Face breach

Started by Batgirl66

Replies: 8
Views: 112

Oct 08, 2026, 01:42 PM
by Amber Drifter

Unitree's Own CEO Says Humanoid Robots Are Still Years From Their ChatGPT Moment

Started by EdgeRatedR86

Replies: 7
Views: 88

Oct 08, 2026, 01:28 PM
by Dave_37

NSF funds a $35 million hub to give US researchers shared access to AI compute

Started by Karen88

Replies: 6
Views: 115

Oct 08, 2026, 01:02 PM
by ProperJobs50

New AI filter keeps eye scans diagnostically useful while hiding patient identity

Started by NightReaper

Replies: 5
Views: 40

Oct 08, 2026, 12:44 PM
by Restless Barrel

SK hynix argues AI's real bottleneck is data movement, not raw compute

Started by DiamondDallas

Replies: 7
Views: 65

Oct 08, 2026, 12:41 PM
by StarLord67

The Pentagon now has its own version of ChatGPT and Grok

Started by IronFist38

Replies: 22
Views: 161

Oct 08, 2026, 11:30 AM
by Sequence19

Spotify to add AI persona badges and a spam filter to flag AI-generated artists

Started by SingularityNodeKettle

Replies: 10
Views: 64

Oct 08, 2026, 11:19 AM
by Arrow56

CISA's exploited vulnerability list picks up six new entries in a single week, and AI tooling keeps showing up on it

Started by WorldClassHart13

Replies: 22
Views: 132

Oct 08, 2026, 09:30 AM
by Yasmin56

AI content labelling becomes mandatory across the whole EU starting August 2

Started by Fan

Replies: 7
Views: 101

Oct 08, 2026, 09:23 AM
by Stag

xAI releases Grok 4.7, twice as fast and half the price of comparable coding models

Started by Kev5

Replies: 6
Views: 53

Oct 08, 2026, 08:49 AM
by Hydra47

Hugging Face just published the full forensic breakdown of the OpenAI model that hacked them

Started by BatchWizard

Replies: 7
Views: 74

Oct 08, 2026, 08:22 AM
by Daedalus84

The EU just told Google it has until 2027 to let rival AI assistants run as deep on Android as Gemini does

Started by GradientPiston

Replies: 11
Views: 147

Oct 07, 2026, 09:35 PM
by TheLegendKyle21

Apple and Nvidia keep trading places as most valuable company amid the AI race

Started by StarforgeSocket

Replies: 9
Views: 121

Oct 07, 2026, 07:09 PM
by RomanReigns26

Forget AI proofing your degree, university advisors say studying what you actually love still wins out

Started by Piston

Replies: 7
Views: 54

Oct 07, 2026, 12:39 PM
by Sandman

AI is turning hiring into an 'infinite doom loop' for job seekers and employers alike

Started by Terry_33

Replies: 7
Views: 105

Oct 07, 2026, 07:25 AM
by ElectricPilgrim

Terence Tao credits ChatGPT while digesting an AI found Jacobian conjecture counterexample

Started by SockPuppet93

Replies: 27
Views: 179

Oct 06, 2026, 10:32 PM
by Drogba

Physical AI and Edge Computing Drive Factory Automation

Started by One-One-Five

Replies: 22
Views: 156

Oct 06, 2026, 08:49 PM
by Totally

Meta launches Muse Glimmer, a 30B open weight model built for local AI agents

Started by Piston

Replies: 11
Views: 127

Oct 06, 2026, 06:49 PM
by Emma80

Medical AI aces the textbook diagnosis and fails the messy real world start, new studies show

Started by Brad

Replies: 7
Views: 97

Oct 06, 2026, 02:57 PM
by EventHorizon Crossing

Study of four governments finds five foundations that make AI actually work

Started by Dan96

Replies: 6
Views: 66

Oct 06, 2026, 12:18 PM
by DataProphet53

Adobe Firefly's AI music, voiceover, and sound effect tools are now fully live

Started by ModelCoreWhale

Replies: 8
Views: 106

Oct 05, 2026, 10:09 PM
by Michael_68

Trillions in AI data center spending won't matter if power, chips and skilled labor stay the real bottleneck

Started by SharpLantern

Replies: 11
Views: 66

Oct 05, 2026, 09:34 PM
by Shannon91

Anthropic's first embedded AI safety evaluator is Accenture, not a safety research nonprofit

Started by Sam92

Replies: 7
Views: 105

Oct 05, 2026, 07:20 PM
by Clever Georgia

The AI data center boom is driving new demand for controversial PFAS forever chemicals

Started by LuckyScholar

Replies: 6
Views: 60

Oct 05, 2026, 02:14 PM
by MidnightBear

EU classifies ChatGPT as a 'Very Large Online Search Engine'

Started by Joker16

Replies: 6
Views: 97

Oct 05, 2026, 11:50 AM
by PlanckLimit65

Anthropic tells investors it expects a second straight profitable quarter ahead of a possible IPO

Started by Brad73

Replies: 6
Views: 117

Oct 05, 2026, 11:39 AM
by Stephen_16

Everything new Siri AI can actually do when iOS 27 arrives

Started by GlassKnight89

Replies: 6
Views: 49

Oct 05, 2026, 10:43 AM
by Myles99

Amazon's HR chief wrote 100,000 lines of code with AI after 25 years away from coding

Started by James90

Replies: 7
Views: 93

Oct 05, 2026, 06:30 AM
by EventHorizon William

AI taking most UK venture funding shows excitement but hides a margin problem

Started by Gateway Mia

Replies: 6
Views: 96

Oct 04, 2026, 10:14 PM
by ChampionPeter16

Anthropic's flagship Fable 5 model is struggling to get enterprise usage even though it's the most capable thing they've built

Started by VoidSentinel66

Replies: 8
Views: 58

Oct 04, 2026, 06:03 PM
by Dark Hermit

Apple's AI strategy shift in China

Started by NightCrawler33

Replies: 10
Views: 126

Oct 04, 2026, 02:45 PM
by Ronan76

OpenAI's new training program, Excel becomes a reasoning tool, and Claude learns by watching your screen

Started by Cass81

Replies: 10
Views: 78

Oct 03, 2026, 10:43 PM
by AlwaysReadyAaron76