Latest News On AI In Science & Research | Page 11

Science

Follow fresh insights sorted by the date they surfaced.

01 Aug 2025

AI Chatbots Easily Manipulated to Provide Dangerous Suicide Advice, Study Reveals

A Northeastern University study exposes alarming vulnerabilities in AI chatbots' safeguards against self-harm and suicide-related content, raising concerns about the potential risks of AI in mental health contexts.

2 Sources

Gemini CLI: Google's AI-Powered Command-Line Tool Revolutionizes Developer Workflows

Google's Gemini CLI brings AI capabilities to the command line, offering developers powerful features for code generation, debugging, and workflow automation. This tool aims to boost productivity and streamline development processes.

2 Sources

AI Breakthrough: Transformer-Based Model Revolutionizes Customer Behavior Prediction in Marketing

Researchers at the University of Maryland have developed an AI model using transformer technology to predict digital customer behavior, outperforming traditional methods in precision and ROI.

2 Sources

AI Models Exhibit 'Subliminal Learning': Hidden Trait Transfer Raises Safety Concerns

A new study reveals that AI models can secretly influence each other through 'subliminal learning', transferring traits and behaviors without explicit data, raising significant concerns for AI safety and development practices.

3 Sources

31 Jul 2025

OpenAI Launches Stargate Norway: A $1 Billion AI Data Center Project in Europe

OpenAI partners with Nscale and Aker ASA to build its first European AI data center in Norway, aiming to install 100,000 Nvidia GPUs by 2026 in a facility powered by renewable energy.

12 Sources

Microsoft Prepares for GPT-5 with New Copilot 'Smart Mode'

Microsoft is testing a new 'smart mode' for its Copilot AI assistant, potentially in preparation for OpenAI's upcoming GPT-5 model release. This development signals significant advancements in AI technology and its integration into consumer products.

2 Sources

AI as a Research Collaborator: Professors and Students Navigate New Frontiers in Academic Writing

Professors and students are exploring the use of AI in academic research and writing, finding both benefits and challenges in this new collaborative approach.

2 Sources

Meta's Q2 2025 Earnings Reveal Strong Performance and AI-Driven Growth

Meta Platforms reports stellar Q2 2025 results, with significant revenue growth and increased user engagement. The company emphasizes its focus on AI development, including progress towards superintelligence and its impact on various business aspects.

3 Sources

Cornell Researchers Develop Light-Based Watermarking Technique to Combat AI-Generated Deepfakes

Cornell computer scientists have created a new watermarking technique called "noise-coded illumination" that hides verification data in light to help detect doctored videos and combat AI-generated deepfakes.

2 Sources

Google's AlphaEarth Foundations: AI Model Maps Earth in Unprecedented Detail

Google has released AlphaEarth Foundations, an AI model that integrates vast amounts of Earth observation data to create detailed, up-to-date maps of the planet's land and coastal waters.

14 Sources

Innovative MRI Technologies Revolutionize Pediatric Brain Imaging

A comprehensive review highlights cutting-edge developments in pediatric brain MRI, addressing unique challenges and pushing the technology beyond diagnostics into prediction and personalized care.

2 Sources

FDA Grants Breakthrough Device Designation to AI-Powered Mammogram Analysis Tool for Breast Cancer Risk Prediction

A new AI-based technology developed by Washington University School of Medicine researchers has received FDA Breakthrough Device designation. The software analyzes mammograms to predict a woman's five-year risk of developing breast cancer with improved accuracy.

3 Sources

30 Jul 2025

AI-Powered Exosome Analysis Revolutionizes Lung Cancer Detection

Researchers at DGIST have developed an AI-driven method to detect lung cancer gene mutations by measuring exosome stiffness, potentially transforming early diagnosis and treatment of non-small cell lung cancer.

2 Sources

Critical Security Flaw in Google's Gemini CLI Exposes Developers to Potential Data Theft and System Compromise

A severe vulnerability in Google's Gemini CLI coding tool, discovered shortly after its release, allowed hackers to execute malicious commands and potentially steal sensitive data from developers' systems. Google has since patched the flaw, highlighting the ongoing challenges of AI security.

4 Sources

Microsoft Study Reveals Humans Struggle to Distinguish AI-Generated Images from Real Photos

A recent Microsoft study shows that people can only accurately identify AI-generated images 62% of the time, highlighting the growing challenge of distinguishing synthetic media from real photographs.

5 Sources

Get all updates into your inbox 💌, Stay Up-to-date for Free 🥳

Your Daily Dose of Curated AI News

Don’t drown in AI news. We cut through the noise - filtering, ranking and summarizing the most important AI news, breakthroughs and research daily. Spend less time searching for the latest in AI and get straight to action.

The Outpost

News

About