How AI Is Upending Politics, Tech, the Media, and More (www.wired.com)

At WIRED’s AI Power Summit in New York, leaders from tech, politics, and publishing laid bare a growing rift over AI’s impact: industry figures hailed AI’s potential for economic growth and scientific progress, while media executives warned that summarization tools are siphoning traffic and revenue from publishers. Participants included a U.S. senator calling for preemptive guardrails around copyright and platform harms, an author of the Trump Administration’s AI Action Plan defending a more regulated approach, and Google’s policy VP touting AI’s benefits in protein modeling and materials science. Publishers countered that “AI Overviews” — which generate summaries and still link to sources — have substantially reduced referrals, prompting companies like Gannett to build their own answer engines (DeeperDive) and calls for compensation analogous to music-streaming licensing. For the AI/ML community this matters technically and economically: disputed training data provenance and downstream monetization could drive legal and regulatory constraints on model training, data access, and evaluation benchmarks. Changes in referral dynamics and publisher pushback may force models to expose provenance, adopt paid data licenses, or shift deployment models (platform vs. publisher-hosted). The summit highlights an emerging policy and business battleground where model builders, dataset owners, and regulators will negotiate access, attribution, and incentives—decisions that will shape research practices, model architectures, and product strategies going forward.