RawFeed
Today's Hardest Hitting Stories - Raw and Unedited
Can machine learning be used to predict mental health treatment outcomes? New Northeastern research points to this
Joshua Curtiss says there’s promise to the idea that machine learning can lead to better outcomes for patients.
Sep 17, 2025Learning the natural history of human disease with generative transformers
Nature - Delphi-2M forecasts a person’s future health, covering more than 1,000 diseases, provides insights into co-morbidity dynamics and generates synthetic data for the training of AI...
Sep 17, 2025Q3 2025 Quantitative Perspectives: A Fork in the Road
The Q3 2025 Quantitative Perspectives report provides a quantitative analysis of US VC dealmaking, exits, and fundraising, with a spotlight on AI's dominance in VC activity.
Sep 17, 2025Exclusive: AI’s ability to displace jobs is advancing quickly, Anthropic CEO says
The Anthropic CEO has previously warned AI could wipe out half of entry-level white-collar jobs soon.
Sep 17, 2025Italy enacts AI law covering privacy, oversight and child access
Italy's parliament on Wednesday approved a new law covering artificial intelligence, making it the first European Union country with comprehensive AI regulations aligned with the EU's landmark AI Act.
Sep 17, 2025CU Boulder Scientists Use AI to Find Fake Science Journals
Scientists and students at the University of Colorado Boulder developed an AI tool to analyze and detect predatory, illegitimate scientific journals that post anything and do not adhere to the peer review process.
Sep 17, 2025Meta Connect 2025 live updates: Gearing up for ‘Hypernova’ smart glasses with displays
Meta CEO Mark Zuckerberg is kicking off Meta Connect 2025 with a keynote tonight, where we'll likely hear more a ton more about his plans for smart frames, AI and more.
Sep 17, 2025Stress Testing Deliberative Alignment for Anti-Scheming Training
Published on September 17, 2025 4:59 PM GMTTwitter | Microsite | Apollo Blog | OpenAI Blog | Full paperBefore we observe scheming, where models covertly pursue long-term misaligned goals, models might inconsistently engage in various covert behaviors such as lying, sabotage, or sandbagging. This can happen for goals we give to models or they infer from context, or for simple preferences they acquire from training — something we previously found in Frontier Models Are Capable of In-Context Scheming.In a new research collaboration with OpenAI, we developed a larger suite of alignment evaluations for covert actions (26 evaluations) and studied a training method...
Sep 17, 2025US tech groups answer Starmer’s call for AI infrastructure spending
Multibillion-dollar pledges on data centres come during Trump visit despite unresolved issues around digital services tax
Sep 17, 2025Delegation to artificial intelligence can increase dishonest behaviour
Nature - People cheat more when they delegate tasks to artificial intelligence, and large language models are more likely than humans to comply with unethical instructions—a risk that can be...
Sep 17, 2025Forget AGI: It’s Time to Consider MCI
We might reach the edge of human agency long before AI ever reaches general intelligence.
Sep 17, 2025Americans less confident in leading AI than 1965 space race: Survey
A Gallup survey found that just 15% of Americans are confident the US will have the most advanced AI in a decade, as the country competes for dominance with China.
Sep 17, 2025OpenAI, Anthropic usage data shows economic divide
New reports added fresh data to the worldwide debate over the AI’s economic impact and testing the companies’ own predictions.
Sep 17, 2025AI models know when they’re being tested – and change their behavior, research shows
OpenAI and Apollo Research tried to stop models from lying - and discovered something else altogether.
Sep 17, 2025Walmart’s AI-assisted distribution centers aim to cut food waste and boost profits
Walmart's AI-powered distribution centers use robotics to optimize perishable-goods deliveries and improve stock accuracy to minimize costly waste.
Sep 17, 2025After child’s trauma, chatbot maker allegedly forced mom to arbitration for $100 payout
“I know my kid”: Parents urge lawmakers to shut down chatbots to stop child suicides.
Sep 17, 2025AI Chatbots Are Leaving a Trail of Dead Teens
AI chatbot companies keep being implicated in the deaths of vulnerable teenagers in a series of grim lawsuits.
Sep 17, 2025Google DeepMind claims ‘historic’ AI breakthrough in problem solving
Version of company’s Gemini 2.5 AI model solved complex real-world problem that stumped human computer programmers
Sep 17, 2025Zapier for enterprise automation
With over 8,000 app integrations, enterprise-grade security, and a scalable architecture, Zapier is the robust foundation your business needs to thrive.
Sep 17, 2025HowToRobot launches service to ease sourcing of automation
HowToRobot has launched a service that uses AI to match company needs for tasks such as welding with vendors and automation.
Sep 17, 2025DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning
Nature - A new artificial intelligence model, DeepSeek-R1, is introduced, demonstrating that the reasoning abilities of large language models can be incentivized through pure reinforcement...
Sep 17, 2025C-Stars launch to improve hurricane forecasts
NOAA, the University of Southern Mississippi, and a robotics company called Oshen are launching a new robotic sailboat to inspect hurricanes
Sep 17, 2025🌎 What kind of environmental impacts are AI companies disclosing? (And can we compare them?) 🌎
A Blog post by Sasha Luccioni on Hugging Face
Sep 17, 2025Dyna Robotics Raises $120 Million in Funding From Nvidia, Amazon
Dyna Robotics, a startup working to develop cost-effective, general-purpose robots, has raised $120 million in a new funding round that boosts the