They hired someone to talk you out of it

OpenAI is hiring a Community Engagement Lead in Ohio. The posting pays $129,600-$236,000 plus equity. The duties include "reduce risk of opposition" and "monitor community sentiment and advise executives on risks and opportunities." A salesman with a nicer title, pointed at the people who live next to the build.

On August 17 the same company locked in about 8 gigawatts of IT capacity at the PORTS-Pike campus in Pike County. The press release talks about 35,000 construction jobs and 2,500 long-term operating jobs. The Community Engagement Lead is who you hire when those numbers are not closing the room.

Heatmap Pro asked 2,045 registered voters, August 8 to 13, whether they would support a data center near where they live. Seventy-five percent said they would oppose it. More than six in ten said they would strongly oppose it. A year ago the same wording was a coin flip, 43% support and 42% oppose. That is a 33-point swing in twelve months, across party, age, income, and zip code. More than 530 counties and towns have already restricted or banned the facilities.

Three-quarters of the country doesn’t want a data factory next to their house. A new slogan won’t make us forget!

Sam Altman sat down with David Senra and treated that poll like a slight branding miss. AI builders have "not, as a field, done a very good job of explaining to people what the benefits are and how the downsides can be mitigated." Then he acted out the old pitch. A 25% chance we destroy the world, and we race anyway so the bad guys don't get there first. Or 50% of the jobs gone next year, "and we hope you all are OK." He even mocked the tone: "dear peasants, we will bequeath upon you these gifts."

People already heard the product. It’s too late! They saw the layoff email that said AI. They saw the empty lot. They formed an opinion. Altman would very much like you to forget.

Anthropic is running the same math. They’re dressing it up differently. People familiar with the company's IPO prep told CNBC that the prospectus will list public backlash against AI and data centers as a risk factor. The safety lab is about to warn Wall Street that Americans noticing the buildout could hurt the stock.

A new Anthropic engineer described the first weeks on Slack like this: everything he does is being "watched," the system surfaces discussions he should join, and it even suggests what he should say.

The product underneath it is first-party. Claude Tag, launched in June, reads every top-level message in the channels it belongs to, tagged or not. Ambient mode means it jumps in when it thinks you need to know something. Anthropic says 65% of its product team's code is already created by the internal version. A later company writeup put Claude-authored merged code above 80%. Fiona Fung, who leads Claude Code, told Lenny Rachitsky the team started feeling lonely because they were talking to agents instead of each other.

The company that lectures the world about control is piloting ambient workplace monitoring on its own engineers first. Your Slack habits, then a suggestion about what to type. I wrote about identity-shaped answers a couple of issues ago. Same deal here.

The Discord from last issue is still the free room. AI, power, work, the show, and the stuff that does not fit a feed.

The move: join the free Dr. J Discord. Read the rules, say hello in #a-general, and watch #a-announce.

If you are in Ohio, or anywhere they want to park 8 gigawatts next to a school, the poll already did the talking. The new hire's job is to talk you back down. Don't make it easy.

- Dr. J