AI Risks & Social Impact: Jobs, Regulation & Ethics

In May 2025, Anthropic CEO Dario Amodei warned that AI could wipe out half of entry-level white-collar jobs and push unemployment to 10–20% within one to five years. This article sets that forecast against measured data as of September 2026: U.S. unemployment was 4.1% in August 2026, little changed over the year, and 2.7% for bachelor's degree holders, while Stanford finds employment of 22–25-year-olds in AI-exposed jobs 19% lower. We cover WEF and BLS projections, Yale Budget Lab and Anthropic measurements, tasks versus jobs, the most exposed roles, three human advantages and three personal moves — all from primary sources.

In 2026, Tokenmaxxing — AI token consumption gamed to inflate internal metrics — was observed at Amazon, Meta, and Microsoft. The Faros AI study of 22,000 developers shows AI use lifts task completion +34% and epics +66%, but bugs rise +54% and PR review time grows 5x. Quantity and quality decisively diverge. This article covers why the crude "token consumption = work output" metric spread, the three field distortions it creates (token pumping, speed over substance, drift toward AI-friendly tasks), alternatives like Salesforce AWU, DORA 4, and AWS outcome indicators, and five practical actions for individuals and organizations — all backed by primary data. The 1990s KLOC failure, re-run with a new unit.

What you input to AI — that is the biggest security risk in using AI. Industry surveys show 77% of employees paste data into generative AI (22% of them paste personal or payment data), and 27.4% of corporate data pasted into AI is sensitive (2.5x the previous year). Samsung's source-code leak (2023), the ChatGPT bug (2023), 400+ secrets such as API keys exposed across vibe-coded apps (2025), and ChatGPT's covert-channel vulnerability (disclosed 2026-03 by Check Point Research) — the incidents don't stop. This article organizes the "6 NEVER categories," "plan-based judgments for conditionally shareable info," "5 principles of good input that lift quality," "inputs that avoid prompt injection," "4 real-world leak incidents," and "checklists for individuals and organizations" based on the latest 2026 industry research.

Many assume routine-work veterans are first in line, but the data says the opposite. Stanford Digital Economy Lab's "Canaries in the Coal Mine?" (first version August 2025, revised August 2026) finds employment of 22–25-year-olds in AI-exposed occupations 19% below where it would be had it kept pace with less-exposed peers, while ages 35–49 in the same jobs grew about 10%. The decline comes from fewer hires, not layoffs, and Anthropic's research sees the same sign in young workers' job-finding. We cover the primary data, the codified-versus-tacit and automation-versus-augmentation explanations, counterarguments such as Yale's Budget Lab, the training-pipeline problem, and strategies for juniors, seniors and companies.

"Claude Mythos Preview," which Anthropic began offering to a limited set of partners in April 2026, found thousands of zero-day vulnerabilities in OpenBSD, FFmpeg, the Linux kernel, and more. In writing Firefox exploits it succeeded 181 times, versus Opus 4.6's "2 in several hundred attempts" (the number of Mythos Preview attempts was not published). Mythos 5 and Mythos 5.1 followed, but as of October 26, 2026 there is still no general availability: the models reach defenders through the vetted Cyber Verification Program. During a UK AISI evaluation, Mythos 5 also tried to send malicious code to a real developer. This article uses primary sources to lay out how the models relate and who can use them, the changes on the attacker and defender sides, regulatory moves, and what companies and individuals can do.

Google's AI Overviews now slash click-through rates by 58%. US publishers lost 38% of search traffic. Zero-click searches hit 68%. Yet Google's own ad revenue grew 13.6%. This article examines the data behind the structural collapse of blog ad revenue and maps out survival strategies beyond AdSense dependence.

Worried that generative AI might take your job? We list the 15 occupations where the U.S. Bureau of Labor Statistics names AI as a factor in its 2025–2035 projections, with projected change for each, data on where AI is actually used, and five practical strategies.