You are currently viewing Tech and AI News for the Week of May 5th, 2025

Tech and AI News for the Week of May 5th, 2025

This week’s key AI developments were underscored by this theme: the candid admission from Anthropic’s CEO that we have no idea how AI works. Regardless, meaningful advances from this remarkable technology continue to accelerate. This week, the FDA announced an aggressive AI rollout speeding up approvals for drugs and medical devices in the US. Conversely, a new study suggesting AI use could damage professional reputation points to an emerging social and professional balancing act as AI adoption broadens. Conversations about AI’s structural impacts on the job market were also a major theme this week. Here are the week’s highlights.

Top Stories

  • BIGGEST STORY – OpenAI is reportedly exploring a new for-profit structure, potentially becoming a Public Benefit Corporation, a move sparking debate and scrutiny, including from Elon Musk, as it navigates its organizational future and financial needs. (link)
  • MOST SIGNIFICANT TECH ADVANCEMENT – Engineers have developed the first programmable photonic chip capable of training nonlinear neural networks using light, a breakthrough promising to significantly accelerate AI training, reduce energy consumption, and advance light-powered computing. (link)
  • BIGGEST REVELATION – Anthropic’s CEO Dario Amodei candidly admitted that experts, including those at leading AI labs, do not fully understand the intricate workings of complex AI models, highlighting a significant knowledge gap in the field. (link)
  • BIGGEST MODEL – Google’s Gemini 2.5 Pro I/O Edition has reportedly surpassed Anthropic’s Claude 3.7 Sonnet as the leading AI for coding tasks, showcasing enhanced capabilities, including building full web apps from a single prompt. (link)
  • BIGGEST POLICY MOVE – The U.S. FDA announced the completion of its first AI-assisted scientific review pilot and an aggressive timeline to scale AI use agency-wide by June 30, aiming to enhance review efficiency. (link)
  • CURVEBALL OF THE WEEK – An AI-generated deepfake clone of a man murdered four years ago delivered a victim’s impact statement in court, a novel and poignant use of technology highlighting its evolving role in the justice system. (link)
  • EDITOR’S PICK – A Duke University study suggests using AI tools at work can negatively impact professional reputation, with colleagues potentially viewing AI users as lazier or less competent, leading some to hide their AI usage. (link)
  • BUSINESS HIGHLIGHT – Most enterprise AI spending is driven by fear of missing out (FOMO) rather than proven return on investment, with only about one in four AI initiatives delivering expected ROI so far, an IBM survey found. (link)
  • MEDIA MOMENT – UK startup Moth and artist ILĀ released “Recurse,” reportedly the first commercial song created using quantum-powered generative AI, showcasing a new frontier in the fusion of music, AI, and quantum computing. (link)
  • CODING ADVANCE – Salesforce is ambitiously targeting 50% of its code to be AI-generated by the end of 2025, signaling a major shift in software development practices and leveraging AI for significant productivity gains. (link)
  • CYBERSECURITY ALERT – Over half of all fraudulent activities are now driven by AI, according to a Feedzai report, with criminals leveraging generative AI for deepfakes, synthetic identities, and sophisticated phishing scams. (link)
  • NEW TOOLS – Figma announced a major update with new AI-powered tools, including Figma Sites for website building and Figma Make for generating designs from prompts, aiming to become an all-in-one product design platform. (link)
  • EDUCATION INNOVATION – A Texas-based school, Alpha School, is utilizing AI as the primary teacher for core academics, compressing learning into two-hour daily sessions, with students reportedly achieving high national scores. (link)
  • ROBOTICS DEVELOPMENT – Amazon unveiled Vulcan, its first AI-powered warehouse robot with a sense of touch, designed to handle a wider variety of items with human-like finesse, marking a leap in warehouse automation capabilities. (link)
  • STARTUP SPOTLIGHT – CodeAnt AI, a Y Combinator-backed startup, raised $2 million to develop its AI-powered platform for automating code reviews, aiming to address the bottleneck created by the rise of AI-generated code. (link)
  • INDUSTRY SHIFT – PwC cut 1,500 U.S. jobs, citing low attrition and a need to align for the future, a move some see as driven by automation and AI rather than recessionary pressures. (link)

Themed Link Stacks

AI in the Workplace & Economy

The Evolving Landscape of AI Models

Societal Impacts & Ethical Concerns

AI Transforming Industries

Deep Dive

A recent Duke University study reveals a “social evaluation penalty” for using AI tools in the workplace, suggesting that while AI can boost productivity, it might harm one’s professional standing. The research, involving over 4,400 participants across four experiments, found that colleagues often perceive individuals who use AI as lazier, less competent, less diligent, less independent, and less self-assured. This stigma was consistent across various demographics, including the target’s age, gender, or occupation. The study also showed that these negative perceptions can translate into real-world consequences, such as a decreased likelihood of being hired by managers who don’t use AI themselves. Interestingly, managers who frequently used AI tools showed a preference for candidates who also used AI. The core reason for this penalty appears to be perceptions of laziness. However, this negative view diminishes when AI benefits the assigned task or when evaluators are themselves frequent AI users. These findings highlight a hidden social cost that could hinder AI adoption within organizations, even as companies push for wider AI implementation. Employees may resist due to concerns about how peers and superiors will perceive their AI usage.

🔗 Full piece

That is it. Stay tuned for more news next week. I APPRECIATE ALL THE READERS. Help spread the word. I put these newsletters together so productive humans like you can stay on top of the latest AI stories shaping our world.

If you like these weekly tech news reports, subscribe to get notified of new editions and updates. For daily updates, check out our news page. For a more in-depth analysis of the week’s news, sign up for our free weekly newsletter to the right of the daily news, or follow me on Twitter or YouTube.

Footer

See the complete archive of 70 stories

Social

Tech Advancements

Business

Media

Law

Coding

Cybersecurity

Hardware

Tools

Education

Government

Robotics

Healthcare

Startup