AI regulation took center stage in San Francisco's 11th District congressional race, where State Sen. Scott Wiener and Supervisor Connie Chan laid out sharply different approaches. Wiener pushed for a federal AI safety body with kill-switch authority over dangerous systems, while Chan called for a moratorium on new data centers and greater worker involvement in oversight decisions. The debate underscored how quickly the policy conversation around AI is moving.
Key Takeaways
- San Francisco congressional candidates disagree on AI regulation, with Wiener favoring a federal safety body and Chan pushing for a data center moratorium.
- Washington state leaders are advancing AI regulation independently, drafting bills on liability, safety audits, and student privacy.
- A 2026 IFS survey found 43 percent of teams deployed AI workers, but only 4 percent require human approval for AI decisions.
- Accenture's Marc Warner said nearly zero companies have truly transformed their core operations around AI.
- Flex CEO Revathi Advaithi Warns Power grid is unprepared for AI growth and estimates a decade-long power gap.
- Wirecutter's hybrid ad and affiliate deals grew over 300 percent year over year as AI changes product search.
- One couple used ChatGPT to audit the mental load in their marriage and rethink household responsibilities.
- Evotec's AI use improved Phase I clinical trial success rates to 80 to 90 percent from 50 to 60 percent.
- Jefferson University launched five AI-focused programs, with President Susan Aldridge citing growing demand for specialized skills.
- Experts say AI safety is an engineering problem requiring risk-based oversight, red-teaming, and drift monitoring.
San Francisco congressional candidates debate AI regulation plans
State Sen. Scott Wiener and San Francisco Supervisor Connie Chan are running for Congress in California's 11th District. They disagree on how to regulate artificial intelligence. Wiener supports a federal AI safety body and kill-switch powers. Chan wants a moratorium on data centers and more worker input. The debate highlights the urgent need for AI rules as development speeds up.
Washington state leaders push for AI regulation without Congress
Federal lawmakers have made little progress on AI rules, so Washington state leaders are acting on their own. Senate Majority Leader Jamie Pedersen and Sen. Patty Murray held roundtables to discuss AI guardrails. State Rep. Clyde Shavers is drafting three bills on AI liability, safety audits, and student privacy. Experts say Washington must also pass data privacy laws before adding AI protections.
Workers train AI colleagues while HR falls behind
Frontline employees are training each other to use AI tools, but HR departments are not keeping up. A 2026 IFS survey found 43% of teams deployed AI workers in the past year. Only 4% of companies require human approval for AI decisions. Experts say HR must formally recognize worker-led AI training and set clear rules for oversight.
Human in the loop is an engineering problem for AI safety
Making AI safe is an engineering task, not just a matter of adding human reviewers. The author says companies should choose oversight modes for each AI action based on risk. They recommend golden data sets, red-teaming, and drift monitoring to test AI systems. The article warns that AI errors range from embarrassing to dangerous depending on context.
Evotec combines human scientists and AI in drug discovery
Evotec, a drug discovery company in Hamburg, uses AI to speed up research but keeps human experts in charge. The company uses AI for target identification, screening, and analyzing omics data. AI has improved Phase I clinical trial success rates to 80-90% from 50-60%. However, AI still needs human guidance to understand complex biology for later trial stages.
CPG Radar Rolls Out AI Tools for Food and Beverage Markets
CPG Radar announced the wider rollout of two connected AI platforms called Signals and LeadLogic. Signals analyzes research and consumer trends to help companies spot new product opportunities. LeadLogic matches ingredient suppliers with brands and sales leads. The Cleveland-based company works with more than 60 companies across North America, Europe and Australia.
Couple Used ChatGPT to Audit the Mental Load in Their Marriage
A writer used ChatGPT to analyze household tasks shared between her and her husband. The AI helped reveal that she handled most of the planning and remembering, even when tasks appeared shared. ChatGPT pointed out that she had become the household's anticipatory system. The couple used the data to rethink how they divide responsibilities.
AI Chiefs Say Few Companies Have Truly Transformed With AI
Marc Warner of Accenture and three other executives spoke at a Fortune event about AI in business. They said the number of companies that have truly rebuilt their core operations around AI is close to zero. Most firms use AI only in areas like marketing or meeting notes. The panel said real gains require redesigning workflows, not just adding AI tools.
Flex CEO Warns Power Grid Must Keep Up With AI Growth
Revathi Advaithi, CEO of supply chain company Flex, warned that America's power grid is not ready for AI growth. She said the grid has barely changed since the time of Thomas Edison. She warned that without faster infrastructure upgrades, AI progress could stall. She estimates a power gap will remain for the next decade.
Wirecutter Expands Beyond Affiliate Revenue as AI Changes Search
Wirecutter, a New York Times-owned product review site, is diversifying its revenue as AI changes how people find products. It launched new franchises like the Hall of Fame Awards and Best New Picks, which drew millions of page views. The company now offers hybrid ad and affiliate deals that grew over 300 percent year over year. Wirecutter also added podcasts and video shows.
Jefferson Launches Five AI-Focused Programs for Health and Data Careers
Jefferson University launched five new AI-focused programs in October 2026. The programs include a MS in Computational Biology and Medicine, a MS in Health Artificial Intelligence, an AI in Health Analytics Leadership Certificate, and two degrees in Business and Data Analytics with AI. University President Susan Aldridge said the programs respond to growing demand for specialized skills in technology and analytics. The Health AI program can be completed in about two years while working full time.
Sources
- San Francisco Is Ground Zero for AI. How Would Congressional Candidates Push for Regulation?
- Washington State Leaders Weigh AI Guardrails
- Workers are training AI colleagues on their own. HR needs to catch up.
- Human in the loop is an engineering problem
- Where Human Insight Meets Artificial Intelligence
- CPG Radar Launches AI-Powered Market Intelligence and Sales Platforms for Food, Beverage and Supplement Industries
- I Ran an AI Audit on My Marriage and the Data Changed Everything
- ‘Close to zero’: 4 AI chiefs say no companies have truly changed in response to AI
- The CEO of supply chain giant Flex says infrastructure and the grid needs to scale with AI—or else ‘Forget all the frontier models’
- Inside Wirecutter’s push to diversify beyond affiliate revenue as AI reshapes discovery
- Jefferson Launches Five AI-Focused Programs to Prepare Students for Careers in Health, Life Sciences, and Data
Comments
Please log in to post a comment.