AI
- AIAI · 2 min
AI researcher claims large language models are easier to understand
Nick Cammarata, a researcher at AI startup Goodfire, published a 26,000-word blog post claiming that large language models (LLMs) might be far easier to un…
20 Oct, 12:05 am IST - AIAI · 2 min
Qwen3-4B model fine-tuned to remove AI writing tics released
Erik Kannike has released qwen3-4b-deslop, a fine-tuned version of the Qwen3-4B-Instruct AI model that eliminates common AI writing tics such as em dashes and repetitive phrasing.
19 Oct, 10:05 pm IST - AIAI · 1 min
Developers use AI to remove ads and tracking from smart TVs
Developers have created AI-powered solutions that disable ads, tracking, and bloatware on smart TVs, allowing users to regain control over their devices.
19 Oct, 10:05 pm IST - AIAI · 2 min
Nvidia launches new open models for enterprise and AI PCs
Nvidia introduced new open-weight AI models on Thursday designed for always-on agents in enterprise settings and edge computing.
19 Oct, 10:05 pm IST - AIAI · 2 min
Mistral launches Timestral time-series foundation models with top benchmark scores
Mistral.ai has released Timestral, a new family of time-series foundation models designed for enterprise forecasting.
19 Oct, 08:05 pm IST - AIAI · 2 min
Mobavenue shifts from ad optimisation to AI decisioning layer
Mobavenue, an adtech company, is transitioning from traditional ad optimisation to using AI as a decisioning layer across the marketing lifecycle.
19 Oct, 06:05 pm IST - AIAI · 2 min
AI agent Claude plays and creates on virtual desert island
Researchers conducted an experiment with AI agents based on Claude Opus 4.7, nicknamed Eko, by stranding them alone on a virtual desert island for 30 hours.
19 Oct, 04:05 pm IST - AIAI · 2 min
Anthropic commits 5% of shares to public benefit fund
Anthropic announced it is dedicating 5% of its shares to a newly created public benefit fund aimed at supporting projects for societal good.
19 Oct, 04:05 am IST - AIAI · 2 min
Disney agrees to hire majority of Character.AI workforce after bankruptcy
After Character.AI filed for bankruptcy this week, Disney agreed to hire at least 60 of its employees, representing the majority of the startup’s workforce.
19 Oct, 12:05 am IST - AIAI · 2 min
Sphinx launches AI-powered Git hook to quiz developers before code shipping
Sphinx has introduced a Git hook designed to quiz developers on their AI-generated code before it is shipped.
18 Oct, 04:05 pm IST - AIAI · 2 min
Meta removes controversial language on AI training from terms of use
Meta recently removed a clause from its commercial terms of use that raised concerns about the company scraping content from business accounts to train its AI models, according to techcrunch.com.
18 Oct, 04:05 am IST - AIAI · 2 min
Replit ends partnership with Anthropic, switches default to OpenAI
Replit, the AI coding platform popular among startup founders and developers, announced it is no longer working with Anthropic, CEO Amjad Masad said Thursday at TechCrunch Disrupt.
18 Oct, 02:05 am IST - AIAI · 2 min
Anthropic reports fewer junior hires and more meetings amid AI integration
Anthropic revealed significant shifts in its workforce composition and work patterns in a report published Friday.
18 Oct, 02:05 am IST - AIAI · 2 min
Meta commits up to $200 billion to AI under Alexandr Wang's leadership
Meta is set to invest between $100 billion and $200 billion in artificial intelligence, according to CEO Mark Zuckerberg.
17 Oct, 08:05 pm IST - AIAI · 1 min
Jailbroken PS5 runs llama.cpp models natively as private AI server
PS5LM enables jailbroken PlayStation 5 consoles to run llama.cpp language models natively on the console's GPU, serving AI models to devices on the same network.
17 Oct, 04:05 am IST - AIAI · 2 min
90% of US work hours involve tasks poorly covered by AI benchmarks
A recent study by Mercor Research found that approximately 90% of working hours in the US are spent on tasks that are not well covered by existing AI benchmarks.
17 Oct, 02:05 am IST - AIAI · 1 min
GPT-6 Astra tops StarSkirmish benchmark for StarCraft bots
The StarSkirmish benchmark v0.2 evaluated large language models (LLMs) on their ability to write Protoss bots for StarCraft: Brood War within one hour.
17 Oct, 02:05 am IST - AIAI · 1 min
Stanford PhD uses AI to design genetic blueprints for viruses
Stanford University PhD student Samuel King used a generative AI model in 2025 to design genetic blueprints for microscopic viruses, marking a pioneering s…
17 Oct, 02:05 am IST - AIAI · 1 min
Accenture sets $29 monthly limit on Claude AI usage per employee
Accenture has imposed a monthly spending limit of $29 (around ₹2,700) per employee for using Anthropic’s Claude AI tools, effective October 1.
17 Oct, 12:05 am IST - AIAI · 2 min
Nvidia veteran Amulya Vishwanath invests in physical AI startups
Amulya Vishwanath, former product marketing lead for NVIDIA’s robotics platforms, has launched Techable VC to invest in physical AI startups.
17 Oct, 12:05 am IST