Google has announced Gemini 2.5, its "most intelligent" model that beats competitors in benchmarks.
Llm RSS
DeepSeek released a new version of its open-source language model, DeepSeek, that can run on consumer hardware.
Character AI is adding a new feature that will provide parents with information about their teens' activity on the platform.
Apple faces a class-action lawsuit for "falsely advertising" its Apple Intelligence features that failed to deliver the promised AI capabilities.
Anthropic has now added a search feature to Claude, something its competitor ChatGPT has had for a while.
It's happened again. ChatGPT has falsely accused someone of a horrific crime, leading to a complaint from the None of Your Business (Noyb) privacy rights group.
Phison has announced new aiDAPTIV+ upgrades and they will soon land inside Maingear's upcoming AI concept laptop.
OpenAI is pushing for looser AI training regulations, urging the U.S. to codify "fair use" and ease restrictions on AI development.
A year after launching Gemma and Gemma 2, Google has announced Gemma 3, which it claims is the "most capable" model that can run on a single GPU or TPU.
Google has announced a new Gemini-powered feature that allows Gmail users to add events to Google Calendar.
Foxconn has announced its own large language model, FoxBrain, that it would use to streamline its manufacturing processes within the company.
Microsoft is also developing its own AI models, which are reportedly competitive with those from OpenAI and Anthropic.
OpenAI has released GPT-4.5, a larger version of GPT-4o, with improved knowledge and writing skills. While it shows modest improvements in benchmarks, it's classified as a medium-risk model.
Amazon has launched Alexa+, a new AI assistant powered by Generative AI. Alexa+ offers enhanced conversational abilities, personalization, and integration with various Amazon services.
Perplexity has released R1 1776, a reasoning model based on DeepSeek's R1 model. R1 1776 has the benefit of not censoring answers China doesn't like.
HP is acquiring Humane, and the company has announced the shutdown date for the Ai Pin, advising users to transfer their data off the platform to prevent data loss.
Microsoft has announced an update for Copilot in Excel that allows it to reference files from Word, PowerPoint, and other apps.
DeepSeek's troubles are not over, as another country has joined the list of countries blocking the app, this time over privacy concerns.
Google has announced a new recall feature for Gemini that allows users to access and review past conversations. This feature is currently rolling out to Gemini Advanced subscribers.
During its Q4 earnings call, Reddit's CEO shared plans to integrate AI-powered answers with its regular search.
A next-gen generative AI model from Chinese tech giant Biadu is expected to arrive in the second half of 2025.
_medium.jpg)
Leaked emails reveal that Meta allegedly torrented massive amounts of pirated ebooks to train its AI models. A new lawsuit details internal concerns and attempts to cover up the activity.
Recent reports suggest that OpenAI's former CTO has been recruiting employees from rival AI firms for her new startup, and now OpenAI co-founder John Schulman is among the latest to join.
The next-gen Amazon Alexa upgrade is reportedly all set to be unveiled by the end of this month at a special event.
During Alphabet's earnings call, CEO Sundar Pichai outlined the company's vision for the future of Search, which prominently features AI.
Google has updated its AI principles and removed its previous promise not to use AI in weapon development.
EFF has responded to a recent bill proposal that, if passed, could lead to jail time or fines for using Chinese AI models like DeepSeek.
Mistral AI has just announced its latest model, a 24 billion parameter model that's comparable to GPT-4o mini and Llama 3.3 70B.
Amid the growing competition from companies like DeepSeek, OpenAI has announced a new partnership with the U.S. National Labs, adding to its growing list of government partners.
DeepSeek, which has been around since 2023, is now experiencing its "ChatGPT moment," causing stock prices for Western companies like Nvidia to drop.
Google's response to ChatGPT, Gemini, is getting an upgrade exclusive to Pixel owners, including features like Deep Research on Pixel and more.
Microsoft and OpenAI are making key changes to their partnership, giving OpenAI the ability to build its own AI infrastructure.
With the latest beta software for iPhone, iPad, and Mac, Apple has temporarily stopped AI summaries for news apps after it was accused of displaying false facts.
Generative AI is trending, and Amazon wants to update Alexa with it. However, a new report has highlighted the challenges Amazon is facing in the process.
OpenAI just keeps improving its flagship product, ChatGPT. The AI giant has announced that paying users will now be able to set reminders and tasks on the platform.
Nearly two years after the Google DeepMind unit was formed, LogaKilpatrick, product lead of Google AI Studio, has announced further expansion.
One of the world's largest "shadow library" was allegedly used to train Llama models, and a new filing now accuses Meta of violating the CDAFA.
The New Year's attack outside the Trump Hotel has drawn media attention, with authorities revealing ChatGPT was used to plan the attack.
Last year's Apple Intelligence error about Luigi Mangione's death has led Apple to announce a fix for its notification summary issues.
Microsoft is investing $80 billion in AI data centers to lead the global AI revolution and meet demand for cloud applications.
Microsoft releases Windows 11 22H2/23H2 KB5053657 with new features and improvements
Windows Photos gets JXL support, improved context menus, OCR web search, and more
Microsoft makes important announcement with new Windows 11 24H2 KB5054148 Dev build
Microsoft begins default-deploying New Outlook for Windows on Microsoft 365
Recommended by 
We're on Mastodon!
We're on Mastodon!
