Brian Solis on Mindshift: Reimagining Customer Experience in the Age of AI

In this episode, we welcome Brian Solis, Head of Global Innovation at ServiceNow, a nine-time best-selling author, keynote speaker, and digital futurist. Starting the conversation, Brian shares the inspiration behind his latest book, “Mindshift“, and emphasizes the importance of empathy and a truly customer-centric mindset. Through examples from Amazon, Disney, and IKEA, Brian illustrates how…

Read More

China’s double pressure on European industry intensifies

European manufacturers face a two-pronged attack from China that has German industry associations warning of deindustrialization: on the one hand, artificially cheap Chinese goods are flooding Europe and, on the other, Beijing has demonstrated its willingness to abruptly cut off access to critical inputs such as rare earths and semiconductors. Alarm intensified in October when…

Read More

These are Science News’ favorite books of 2025

Our staff’s favorite books of the year reflected on the role of science in some of society’s most pressing problems, from artificial intelligence to childhood trauma to river restoration. Did we miss your favorite? Let us know at feedback@sciencenews.org. rehabilitation Walter ShoshanaSimon and Schuster | $29.99 In a journalist’s exposé of America’s drug treatment centers,…

Read More

The Counterintuitive Path to AI Adoption – O’Reilly

The following article originally appeared in Gradient flow and is republished here with the author’s permission. We are experiencing a peculiar moment in the development of AI. On the one hand, the demonstrations are spectacular: agents who reason and plan with apparent ease, models that compose original songs from a text messageand research tools that…

Read More

Researchers isolate memorization from problem solving in AI neural networks

Math operations and closed-book data retrieval shared paths with memorization, dropping from 66 to 86 percent of performance after editing. The researchers found that arithmetic was particularly fragile. Even when the models generated identical chains of reasoning, they failed in the calculation step after low-curvature components were removed. Figure 3 from the article “From memorization…

Read More