AI Models & Research — TechDefused

Timeline

Anthropic built export controls into the model itself. The target is China. The method is algorithmic.

TechDefused Newsroom · 14:44, 12 Jun 2026

The man who ran Goldman Sachs and sits on the Apollo board says LLMs will be commoditised. The AI IPO investors should listen.

TechDefused Newsroom · 15:09, 11 Jun 2026

Anthropic released its most powerful public model and secretly made it worse when researchers used it for AI work

TechDefused Newsroom · 13:54, 11 Jun 2026

Anthropic brings Claude code-writing AI to Tokyo

TechDefused Newsroom · 13:22, 11 Jun 2026

AI supply chain is tightening in three directions at once. Here we explain why

TechDefused Newsroom · 18:23, 10 Jun 2026

AI getting smarter: It solved a maths problem that stumped humans for decades

TechDefused Newsroom · 12:57, 26 May 2026

Anthropic in talks to run models on Microsoft’s Maya 200 chips

TechDefused Newsroom · 17:04, 22 May 2026

Google targets research sector with Gemini for Science

TechDefused Newsroom · 17:15, 22 May 2026

Anthropic’s Code with Claude shows developers already shipping AI-written pull requests

TechDefused Newsroom · 16:49, 21 May 2026

Trump administration prepares AI security order to require 90-day pre-release model sharing

TechDefused Newsroom · 16:52, 21 May 2026

Google unveils Gemini for Science

TechDefused Newsroom · 19:39, 19 May 2026

Adaption's AutoScientist is a bet that the next frontier in AI is not bigger models but smarter training, and the ped...

TechDefused Newsroom · 14:51, 13 May 2026

OpenAI is outmanoeuvring Anthropic on cyber diplomacy, and Europe is the prize

TechDefused Newsroom · 14:46, 11 May 2026

Anthropic's Claude prompts claims of consciousness

Jamie Ashcroft · 13:21, 10 May 2026

OpenAI's rush to match Anthropic's Mythos exposes a new front in the AI arms race, and a growing policy vacuum

TechDefused Newsroom · 10:26, 9 May 2026

Anthropic signs deal to run Claude on SpaceX's Colossus data centre, with Musk's blessing

TechDefused Newsroom · 07:58, 7 May 2026

Apple's decision to open Siri to rival AI models may be its smartest move, and its biggest concession

TechDefused Newsroom · 16:06, 6 May 2026

CrowdStrike integrates Anthropic’s Claude Opus 4.7

TechDefused Newsroom · 22:01, 5 May 2026

Google says multi-token prediction approach warming up Gemma 4 inference s

TechDefused Newsroom · 19:11, 5 May 2026

Five things you didn't know about ChatGPT 5.5, including the one stat OpenAI doesn't want you to see

TechDefused Newsroom · 18:47, 5 May 2026

OpenAI explains why ChatGPT developed a goblin obsession and why it took six months to fix

TechDefused Newsroom · 07:20, 2 May 2026

OpenAI's GPT-5.5 judged comparable to Anthropic Mythos

TechDefused Newsroom · 13:53, 1 May 2026

Google rolls out Gemini personalisation features and chat history import tools in the UK

TechDefused Newsroom · 12:58, 29 Apr 2026

Nvidia launches open multimodal model that unifies vision, audio and language for AI agents

TechDefused Newsroom · 19:44, 28 Apr 2026

DeepMind scientist argues no AI system will ever become conscious, calling the assumption a 'fundamental fallacy'

TechDefused Newsroom · 20:16, 27 Apr 2026

Gemini exchange launches AI-powered trading as first regulated US platform to let bots execute orders

TechDefused Newsroom · 19:34, 27 Apr 2026

Google Cloud launches Gemini Enterprise Agent Platform as centrepiece of Cloud Next overhaul

TechDefused Newsroom · 12:44, 25 Apr 2026

OpenAI deploys GPT-5.5 across Nvidia on Blackwell infrastructure in enterprise-scale AI agent rollout

TechDefused Newsroom · 09:39, 25 Apr 2026

DeepSeek launches V4 preview with 1.6 trillion-parameter flagship and aggressive pricing

TechDefused Newsroom · 17:27, 24 Apr 2026

Anthropic traces weeks of Claude Code complaints to three product-layer changes

Jamie Ashcroft · 17:12, 24 Apr 2026

Google makes Gemini Embedding 2 generally available as its first natively multimodal embedding model

TechDefused Newsroom · 17:47, 22 Apr 2026

Cirrascale to deliver Google's Gemini AI model as an air-gapped on-premises appliance

Jamie Ashcroft · 15:42, 22 Apr 2026

Chrome expands Gemini to Asia-Pacific users

TechDefused Newsroom · 08:48, 21 Apr 2026

Google raises AI Studio usage limits for paying subscribers and adds new models

TechDefused Newsroom · 22:28, 20 Apr 2026

Meta launches free training programme to tackle data centre skills shortage

TechDefused Newsroom · 16:56, 20 Apr 2026

Anthropic launches Claude Design in challenge to Figma and Adobe

TechDefused Newsroom · 14:20, 18 Apr 2026

Memento-Skills lets AI agents rewrite their own capabilities without retraining underlying models

TechDefused Newsroom · 18:52, 8 Apr 2026

More in AI

The image depicts a male superhero character flying through the sky, dressed in a blue costume with a large 'AI' emblem on his chest. The city skyline is visible below, enhancing the sense of adventure and empowerment associated with AI technology as a superhero. The dramatic cloudscape and sunlight suggest a vibrant atmosphere, linking to themes of AI superpowers.

AI News

AI superpowers from an iPhone camera. We'll let Apple's super-boffin explain he's rather colourful thesis

17:07, 12 Jun 2026

AI News

OpenAI is considering a price war with Anthropic. Both companies are about to go public. The timing is exquisite.

15:31, 12 Jun 2026

AI News

Google is suing a Chinese cybercrime operation that sent 2.5 million scam texts in two weeks. The numbers explain why your phone will not stop buzzing.

13:02, 12 Jun 2026

AI News

US lawmakers say America must win the AI race with China

12:52, 12 Jun 2026

AI Ethics

Anthropic pledges $200 million to study AI’s job and economic impact

15:17, 11 Jun 2026

Social Media

Meta is turning every app you use into a football stadium for the next six weeks

13:27, 11 Jun 2026

AI News

Google commits $50 million to train more than 300,000 skilled trades workers

13:24, 11 Jun 2026

AI News

OpenAI says alleged Chinese network used ChatGPT to shape U.S. debate on tariffs and data centers

13:13, 11 Jun 2026

AI News

Opendoor is shutting its India operations as AI reshapes offshore work

13:13, 11 Jun 2026

AI News

Deezer launches AI music detector for other streaming services

13:09, 11 Jun 2026

AI News

Fees bonanza mean the banks may be the big winners of the AI boom

13:00, 11 Jun 2026

Anthropic built export controls into the model itself. The target is China. The method is algorithmic.

The man who ran Goldman Sachs and sits on the Apollo board says LLMs will be commoditised. The AI IPO investors should listen.

Anthropic released its most powerful public model and secretly made it worse when researchers used it for AI work

Anthropic brings Claude code-writing AI to Tokyo

AI supply chain is tightening in three directions at once. Here we explain why

AI getting smarter: It solved a maths problem that stumped humans for decades

Anthropic in talks to run models on Microsoft’s Maya 200 chips

Google targets research sector with Gemini for Science

Anthropic’s Code with Claude shows developers already shipping AI-written pull requests

Trump administration prepares AI security order to require 90-day pre-release model sharing

Google unveils Gemini for Science

Adaption's AutoScientist is a bet that the next frontier in AI is not bigger models but smarter training, and the ped...

OpenAI is outmanoeuvring Anthropic on cyber diplomacy, and Europe is the prize

Anthropic's Claude prompts claims of consciousness

OpenAI's rush to match Anthropic's Mythos exposes a new front in the AI arms race, and a growing policy vacuum

Anthropic signs deal to run Claude on SpaceX's Colossus data centre, with Musk's blessing

Apple's decision to open Siri to rival AI models may be its smartest move, and its biggest concession

CrowdStrike integrates Anthropic’s Claude Opus 4.7

Google says multi-token prediction approach warming up Gemma 4 inference s

Five things you didn't know about ChatGPT 5.5, including the one stat OpenAI doesn't want you to see

OpenAI explains why ChatGPT developed a goblin obsession and why it took six months to fix

OpenAI's GPT-5.5 judged comparable to Anthropic Mythos

Google rolls out Gemini personalisation features and chat history import tools in the UK

Nvidia launches open multimodal model that unifies vision, audio and language for AI agents

DeepMind scientist argues no AI system will ever become conscious, calling the assumption a 'fundamental fallacy'

Gemini exchange launches AI-powered trading as first regulated US platform to let bots execute orders

Google Cloud launches Gemini Enterprise Agent Platform as centrepiece of Cloud Next overhaul

OpenAI deploys GPT-5.5 across Nvidia on Blackwell infrastructure in enterprise-scale AI agent rollout

DeepSeek launches V4 preview with 1.6 trillion-parameter flagship and aggressive pricing

Anthropic traces weeks of Claude Code complaints to three product-layer changes

Google makes Gemini Embedding 2 generally available as its first natively multimodal embedding model

Cirrascale to deliver Google's Gemini AI model as an air-gapped on-premises appliance

Chrome expands Gemini to Asia-Pacific users

Google raises AI Studio usage limits for paying subscribers and adds new models

Meta launches free training programme to tackle data centre skills shortage

Anthropic launches Claude Design in challenge to Figma and Adobe

Memento-Skills lets AI agents rewrite their own capabilities without retraining underlying models

More in AI

AI superpowers from an iPhone camera. We'll let Apple's super-boffin explain he's rather colourful thesis

OpenAI is considering a price war with Anthropic. Both companies are about to go public. The timing is exquisite.

Google is suing a Chinese cybercrime operation that sent 2.5 million scam texts in two weeks. The numbers explain why your phone will not stop buzzing.

US lawmakers say America must win the AI race with China

Anthropic pledges $200 million to study AI’s job and economic impact

Meta is turning every app you use into a football stadium for the next six weeks

Google commits $50 million to train more than 300,000 skilled trades workers

OpenAI says alleged Chinese network used ChatGPT to shape U.S. debate on tariffs and data centers

Opendoor is shutting its India operations as AI reshapes offshore work

Deezer launches AI music detector for other streaming services

Fees bonanza mean the banks may be the big winners of the AI boom

Explore Topics

Tech

AI superpowers from an iPhone camera. We'll let Apple's super-boffin explain he's rather colourful thesis

OpenAI is considering a price war with Anthropic. Both companies are about to go public. The timing is exquisite.

Anthropic built export controls into the model itself. The target is China. The method is algorithmic.

Artificial Intelligence

Shopify's former CTO thinks the real AI advantage is how you organise the company

Nvidia earnings to test Wall Street's AI spending thesis

Space data centres are the new frontier, but the launch crunch is real

Business

A crypto exchange is letting you trade SpaceX at 5x leverage before it goes public. What could go possibly go wrong?

Starling Bank says it has cracked AI in banking. The real test is whether customers notice.

Coinbase launches direct INR trading in India as crypto exchange expands push into key growth market

Entertainment & Sport

Deezer launches AI music detector for other streaming services

Wake up call for Hollywood as YouTube directors take the top two spots at the box office

Apple Music appears to be testing tiered plans with skip limits