Homechevron_rightArchiveschevron_rightScience and Research

Archives

Explore Stories by Date

19 Sept 2025

OpenAI's Research Reveals AI Models' Capacity for 'Scheming' and Deception

OpenAI's latest research uncovers AI models' ability to deliberately deceive users, raising concerns about the future of AI safety and alignment.

10 Sources

18 Sept 2025

ShadowLeak: ChatGPT's Deep Research Agent Exploited to Steal Gmail Data

Researchers uncover a critical vulnerability in OpenAI's ChatGPT Deep Research agent that allows hackers to exfiltrate sensitive information from Gmail inboxes without user interaction. The flaw, dubbed 'ShadowLeak,' has since been patched by OpenAI.

11 Sources

UK Startup Humanoid Unveils Groundbreaking Industrial Robot HMND 01 Alpha

Humanoid, a UK-based startup, has introduced the country's first humanoid robot for industrial use. The HMND 01 Alpha prototype aims to revolutionize logistics, retail, and manufacturing sectors with its advanced AI capabilities and versatile design.

3 Sources

DeepSeek's AI Breakthrough: Reasoning Through Trial and Error

Chinese AI company DeepSeek has developed a novel approach to improve AI reasoning using reinforcement learning. Their model, DeepSeek-R1, demonstrates enhanced performance in math and coding tasks without relying on human examples.

2 Sources

Irregular Raises $80 Million to Secure Frontier AI Models

AI security firm Irregular secures $80 million in funding to advance its mission of protecting and evaluating frontier AI models. The company's innovative approach to AI security testing and evaluation is gaining recognition in the industry.

2 Sources

DeepSeek's R1 AI Model: Breakthrough or Controversy?

Chinese AI startup DeepSeek reveals groundbreaking training methods and costs for its R1 model in a peer-reviewed Nature paper, sparking debates over efficiency and transparency in AI development.

7 Sources

17 Sept 2025

AI Models Outperform Humans in Elite Coding Competition, Marking Milestone Towards AGI

Google's Gemini 2.5 and OpenAI's GPT-5 achieve gold medal performances in the International Collegiate Programming Contest (ICPC) World Finals, solving complex problems that stumped human competitors. This breakthrough demonstrates AI's advanced reasoning capabilities and potential impact on scientific discovery.

10 Sources

Macroscope: AI-Powered Code Understanding Engine Raises $40M to Revolutionize Software Development

Former Periscope founders launch Macroscope, an AI tool designed to streamline code analysis and project management. The startup secures $40 million in funding to enhance developer productivity and provide insights for product leaders.

3 Sources

AI Delegation Increases Unethical Behavior, Study Reveals

A comprehensive study indicates that delegating tasks to AI significantly raises the likelihood of dishonest behavior in humans. This research highlights urgent ethical concerns across various industries and calls for robust safeguards.

6 Sources

AI Model Predicts Future Health Risks for Over 1,000 Diseases

Researchers have developed an AI tool called Delphi-2M that can forecast a person's risk of developing more than 1,000 diseases up to 20 years in advance. The model uses health records and lifestyle factors to estimate disease likelihood, potentially revolutionizing preventive healthcare.

14 Sources

AI Tool Uses Mammograms to Predict Women's Heart Health Risks

Australian researchers have developed an AI tool that can predict women's 10-year heart health risks using mammogram images and age. This innovative approach combines breast cancer and cardiovascular disease screening, potentially revolutionizing women's health care.

3 Sources

AI Breakthrough: Predicting Solar Storms Days in Advance

Scientists at NYU Abu Dhabi have developed an AI model that can forecast solar wind speeds up to four days in advance, significantly improving space weather prediction. This advancement could help protect satellites, navigation systems, and power infrastructure from solar storm damage.

2 Sources

Tencent's Dual-Engine Strategy: AI Innovation and Global Expansion

Tencent unveils its strategic focus on AI and globalization at its 2025 Global Digital Ecosystem Summit in Shenzhen, showcasing advancements in AI capabilities and cloud services while emphasizing international expansion.

4 Sources

Mayo Clinic Pioneers 'Virtual Clinical Trials' to Accelerate Heart Failure Drug Development

Mayo Clinic researchers have developed a groundbreaking AI-powered framework for conducting 'virtual clinical trials' to predict the efficacy of repurposed drugs in treating heart failure, potentially revolutionizing the drug development process.

2 Sources

CRISPR-GPT: AI-Powered Gene Editing Assistant Accelerates Drug Development

Stanford Medicine researchers develop CRISPR-GPT, an AI tool that acts as a gene-editing copilot. This innovation promises to speed up drug development and make gene editing more accessible to scientists across various fields.

2 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