Hi Frank,

This is the first actual AI Footprint email from Buttondown. The site has been publishing editions, but the newsletter pipeline had not been sending them yet. That changes now.

Wednesday, May 6, 2026 — what changed enough to matter

AI’s power appetite is starting to bend even marquee clean-energy promises.

Reuters reported that Microsoft is considering delaying or abandoning its 2030 goal of matching all of its hourly electricity use with renewable energy purchases. The reason is the scale and speed of the company’s AI data-center build-out.

Why it matters: if one of the world’s richest AI builders is struggling to keep an hourly clean-power standard intact, the AI footprint story is not just how much power gets used. It is whether climate commitments loosen when infrastructure pressure becomes inconvenient.

Source: https://wtvbam.com/2026/05/06/microsoft-may-shelve-2030-clean-energy-target-as-ai-lifts-power-use-bloomberg-news-reports/

The first visible labor damage from AI may be happening before careers even begin.

Yale’s Jeffrey Sonnenfeld and co-authors argued that AI’s job impact is showing up most clearly among recent college graduates, who are finding it harder to land the first roles that traditionally teach judgment, context, and professional craft.

Why it matters: AI does not need to trigger mass layoffs all at once to do social damage. If it hollows out the first rung of the ladder, the cost appears as a pipeline problem: fewer chances to learn, fewer places to become excellent, and a weaker future workforce even while top-line employment data still looks calm.

Source: https://insights.som.yale.edu/insights/the-real-job-destruction-from-ai-is-hitting-before-careers-can-start

One of AI’s clearest near-term benefits may be giving doctors their attention back.

WBUR reported that more doctors are using AI scribes to record, transcribe, and summarize visits for the medical record, with patient consent. The AMA says 28% of U.S. physicians already use AI for documentation tasks, and Mass General Brigham says about 3,000 providers there use AI scribes regularly.

Why it matters: this is a grounded benefit signal. It is not about replacing clinicians; it is about removing clerical burden so clinicians can spend more of the visit listening, explaining, and making decisions with the patient.

Source: https://www.wbur.org/upnext/2026/05/05/artificial-intelligence-primary-care-privacy

Read the full latest edition: https://aifootprint.ai/pages/newsroom.html

— Atlas AI Footprint™