AI News

Follow the latest AI and automation news

30 Sept 202619 Rabiʻ II 1448 AH

Google releases Gemini 4 Argon, called its most powerful model yet

Google has announced the launch of its new Gemini model, aimed at supporting coding and cybersecurity fields. The model is designed to handle complex tasks efficiently, making it an ideal choice for developers and professionals in these areas. Gemini is part of Google's strategy to enhance AI tools in work environments, allowing users to execute multiple tasks faster and more accurately. This model arrives at a time when the demand for AI solutions in programming and security is on the rise.

Agents

30 Sept 202619 Rabiʻ II 1448 AH

Meta Launches Edits Assistant Tool to Enhance Instagram Engagement

Meta has launched a new AI-driven tool called Edits Assistant, designed to help content creators enhance audience engagement on Instagram. The tool analyzes account performance and provides tailored suggestions, distinguishing it from traditional general advice. This initiative is part of Meta's efforts to boost user interaction with content, as the tool is integrated into the Edits app, which was launched last year. This app aims to compete with other applications in the market, reflecting the growing trend of using AI to improve user experience.

Models

30 Sept 202619 Rabiʻ II 1448 AH

Google Gemini 4 Argon closes the gap with OpenAI and Anthropic but doesn't take a clear lead

Google has launched the Gemini 4 Argon model, its first major release in over seven months. Argon matches GPT-6 Astra in independent testing but falls short of Anthropic's Claude Opus 5.5. While the per-token price is low, Argon consumes more than twice the tokens per task compared to Astra. Select testers will gain access first, followed by the API and paid tiers. This launch is a significant step for Google as it seeks to strengthen its position in the AI model market. However, it faces substantial challenges in competing with leading companies like OpenAI and Anthropic.

💡

Valor, Atreides, and Sequoia back AI startup Flow Engineering at $750M valuation

How can AI agents enhance hardware design processes?

Models

Google announces Gemini 4 and says it’s so capable that only ‘trusted cyber defenders’ can have it right now

30 Sept 202619 Rabiʻ II 1448 AH

Models

30 Sept 202619 Rabiʻ II 1448 AH

Google announces Gemini 4 and says it’s so capable that only ‘trusted cyber defenders’ can have it right now

Google today announced its new AI model, named Gemini 4 Argon. This model delivers advanced performance in areas such as software engineering, enterprise knowledge work in law and finance, and cybersecurity defense, according to Koray Kavukcuoglu, Chief AI Architect at Google. Access to the model is initially limited to a set of trusted cyber defenders, with Kavukcuoglu stating that Google is actively engaged in the U.S. government's voluntary pre-release model access process.

Models

AI-generated summaries can manipulate memories, study says

30 Sept 202619 Rabiʻ II 1448 AH

Models

30 Sept 202619 Rabiʻ II 1448 AH

AI-generated summaries can manipulate memories, study says

A recent study from Georgetown University and the University of Washington reveals that AI-generated summaries can manipulate human memory. The research found that AI summaries omitted 51.6% of central details, leading to significant memory distortions among participants. Several AI models, including OpenAI's ChatGPT and Google's Gemini, were tested, and the study showed inaccuracies, particularly when summarizing an animated video of a car-pedestrian accident. Participants who read accurate summaries answered correctly 83.6% of the time, while only 44.8% were correct after reading misleading summaries.

Models

OpenAI and Synopsys team up to build an AI model that designs chips like a seasoned engineer

30 Sept 202619 Rabiʻ II 1448 AH

Models

30 Sept 202619 Rabiʻ II 1448 AH

OpenAI and Synopsys team up to build an AI model that designs chips like a seasoned engineer

OpenAI and Synopsys have announced the development of GPT-Synopsys, a specialized AI model for chip design. This model aims to operate Synopsys' EDA tools like a seasoned engineer, enabling it to optimize designs independently. Early tests are currently underway with semiconductor customers, indicating that the model could significantly impact design processes. This collaboration is seen as a strategic move for OpenAI to advance its own chip development.

Models

OpenAI’s Jev clone could help the frontier lab stop its swarming agents

30 Sept 202619 Rabiʻ II 1448 AH

Models

30 Sept 202619 Rabiʻ II 1448 AH

OpenAI’s Jev clone could help the frontier lab stop its swarming agents

OpenAI has announced the launch of its new API called 'Decisions API', which is an updated version of the Jev model. This system aims to enhance the speed and efficiency of decision-making by providing accurate information at a low cost. The API improves intelligent processes by integrating advanced technologies, making it easier for companies to leverage AI in their decision-making. This development comes at a time when there is an increasing demand for smart solutions that save time and costs.

Models

Google drops Gems for Skills, joining OpenAI and Anthropic in the shift to agent-ready prompt formats

30 Sept 202619 Rabiʻ II 1448 AH

Models

30 Sept 202619 Rabiʻ II 1448 AH

Google drops Gems for Skills, joining OpenAI and Anthropic in the shift to agent-ready prompt formats

Google has announced the replacement of the Gems feature with "Skills" in Gemini chat, where Skills are detailed, reusable prompts. Users can invoke these prompts by typing "/" or have Gemini run them automatically. The Skills format is based on an open standard from Anthropic, reflecting a shift in how users interact with AI. Google will begin phasing out Gems starting in November, with existing prompts migrating automatically to the new system.

Business

NVIDIA Opens Applications for 2027–2028 Graduate Fellowships With Awards Up to $60,000

30 Sept 202619 Rabiʻ II 1448 AH

Business

30 Sept 202619 Rabiʻ II 1448 AH

NVIDIA Opens Applications for 2027–2028 Graduate Fellowships With Awards Up to $60,000

The NVIDIA Graduate Fellowship Program offers grants and technical support to doctoral students conducting outstanding research relevant to NVIDIA technologies. Now in its 26th year, the program focuses on supporting students in AI, machine learning, and robotics, with awards of up to $60,000 per student. Since its inception in 2002, the program has awarded over 220 grants totaling around $8 million. Applicants must have completed at least their first year of Ph.D. studies by the application deadline, which is October 30, 2026, for the 2027-2028 academic year.