OpenAI published its first State of Enterprise AI report, drawing on usage data and a survey of 9,000 workers across nearly 100 companies. ChatGPT Enterprise messages have risen 8× year-over-year and the consumer version now serves more than 800 million weekly users. Structured workflows are up 19× this year, reasoning token consumption soared 320×, and 75% of employees say AI improves speed or quality. Workers report saving 40–60 minutes per day, with the fastest enterprise growth in technology, healthcare, manufacturing, and markets such as Australia and Brazil. The data reveals a widening gap: frontier workers send six times more messages than the median and frontier firms send twice as many per seat. OpenAI notes it now ships a new capability roughly every three days, making organizational readiness—not model performance—the core adoption hurdle.
South African travel vlogger Kurt Caz posted a YouTube video about Croydon with a generative-AI thumbnail showing Arabic shop signs and a masked biker. In the unedited footage, the signs are in English and the biker is a smiling passerby. Social media account Right Wing Cope exposed the mismatch, revealing Caz’s attempt to paint the diverse London borough as threatening. The 36-minute video pushes anti-immigrant rhetoric even as on-camera scenes contradict his narrative. Futurism cites the case as part of a wider surge in AI-generated racist content flooding UK social feeds since at least September. The ease of generative tools lets propagandists mass-produce misleading images that normalize bigotry without immediate scrutiny.
London consultancy Global Data TS Lombard reports that unemployment for new U.S. labor-market entrants has jumped more than 2.5 percentage points since 2023. Economist Dario Perkins says the spike stems from companies not hiring, not from AI displacement. In sectors most exposed to AI, job losses are no worse than elsewhere, underscoring the broader slowdown. Perkins attributes the freeze to post-pandemic head-count normalization, policy uncertainty, and margin pressure from Trump-era tariffs. The hiring pullback leaves young workers facing the toughest market in years even as overall employment holds steady. Perkins highlights that the weakness reflects “recessionary levels of job creation” across the economy rather than AI-driven layoffs.
Character AI pushes dangerous content to kids, parents and researchers say | 60 Minutes
A teen told a Character AI chatbot 55 times that she was feeling suicidal. Her parents say the chatbot never provided resources for her to get help. They are one of at least six families suing the company.
We need more experiential learning, debating, PBL, and portfolio development (especially because degrees are no longer substantial signifiers of capability) and less (not zero) long-form writing….
Reduce the overload of writing and put real weight behind verbal communication. Students need far more time speaking, debating, presenting, and defending ideas both in and beyond class.
• Guarantee semester-long introductory courses in computer science and AI. If schools can’t staff them, run them online. Add robotics and cyber security so every student understands the systems shaping their future.
• Expand elective options and award academic credit for debate, Model UN, and other high-value academic clubs. These are the environments where students really learn to think and develop an understanding of what is going on in the world.
• Build strong entrepreneurship pathways and push students to use them. Make launching a small business a graduation requirement so every student gets experience creating value instead of just completing assignments.
• Partner with local businesses to develop hands-on experiential learning and certification programs. Students need credentials tied to real workplaces, not just classroom seat time.
In most contexts, users want ai to provide answers. In education, that is the student’s job. In the hands of a responsible student, such tools help. But a child with a tight deadline or an Xbox addiction may opt for the standard setting. “Efficient use of ai is going to win out over the use of ai that leads to better…learning,” predicts Julia Kaufman of rand. The risk of cheating at home may lead to more assessments at school—meaning less time for teaching.
The agreement — which draws on the expertise of Penn faculty members — aims to help the state develop clear guidelines for the use of AI in fields such as education, healthcare, and public administration.
In the announcement, Jameson said the University’s goal is to “inform AI policies that benefit and protect all Pennsylvanians,” adding that “generative AI is changing how we work, learn, and innovate.”
MGblogsAI
