Steve Little

Sam Altman says OpenAI shares Anthropic's red lines in Pentagon fight But he still wants a Pentagon deal.

UPDATE: Sam Altman says OpenAI draws the same red lines as Anthropic: no AI for mass surveillance or autonomous weapons. But he still wants a Pentagon deal. Employees at OpenAI and Google signed a joint letter backing Anthropic's stance. www.axios.com/2026/02/27/a...

Steve Little (@digitalarchivist) Anthropic Refuses Pentagon AI Demands "Statement from Dario Amodei on our discussions with the Department of War" What this is about — quickly On February 26, 2026, Anthropic CEO Dario Amodei publi...

Claude is already embedded in classified military networks. The standoff raises a core question: who sets ethical limits on AI in war—the companies that build it or the government that wields it?
substack.com/@digitalarch...

Anthropic CEO draws two red lines with the Pentagon: no mass domestic surveillance, no fully autonomous weapons. The Dept. of War threatens retaliation—supply chain blacklist, Defense Production Act. Anthropic won't budge. This may define how military AI is governed for years.

From Spreadsheet to Research Report: Using Claude’s Custom Skills for Genealogy Claude.ai has been one of my favorite apps for genealogical writing for the past couple years. When I first started experimenting with Claude, I was curious if it could transform my spreadsheet res…

Claude's Custom Skills can turn a genealogy research-log spreadsheet into a footnoted research report. Set up a reusable skill with one prompt, upload your CSV/XLSX, and get polished prose — far faster than the chain-of-thought method from a year ago. familylocket.com/from-spreads...

Claude Skills are the next iteration or evolution of prompt-based genealogy work. Nicole Dyer is a Jedi Master of explaining this technology in a genealogical context.

Loathsome Jargon: Hard Take-Off What the Scariest Term in AI Actually Means

Loathsome Jargon #3: We traced "hard take-off" from a 1965 math paper to this month's AI panic. Every source named. Every claim traceable. The genealogist's method, applied.

vibegenealogy.ai/p/loathsome-...

A karaoke company erased $17.4 billion from the stock market. Look at this chart — AI capability just went vertical. "Hard take-off" is the scariest term in AI right now, and your feeds are full of it. Here's what it actually means and who's saying it.

Maybe stack or wrap the OCR/HTR prompt with this one, a project I've been helping the IAJGS for about 18 months:
github.com/DigitalArchi...

I'd be interested if this prompt with Gemini 3 Pro yields better results. (I was with LVA's Digital Library Program in the late 1990s/early 2000s, working out an annex in the Portsmouth Public Library, digitizing local library historical photograph collections.)

github.com/DigitalArchi...

It's been my observation that accuracy is proportional to legibility; how challenging were the scraps?

I love libraries, archives, and cemeteries too—I'm a genealogist and I train professional researchers to use these tools. AI doesn't replace research. It helps me prepare before I go and organize what I find. There are real problems with this tech and I talk about them openly. It's not either/or.

My personal belief is that time is best spent restoring and strengthening democracy so that we can legislate, regulate, and tax. This tech was built on six thousand years of human culture, and all should benefit.

There are a MESS of problems with this tech--from its creation, through its use, and on to its outputs and impacts--but vague handwaving is not helpful. These problems will best be addressed by dividing-and-conquering, understanding each issue and addressing each specifically.

Steve Little (@digitalarchivist) 🎧 Two Pods on AI's Actual Environmental Footprint Andy Masley — EA DC director, former physics teacher, not an expert but arguably the most careful lay analyst on this topic — spent months fact-check...

Andy Masley fact-checked the viral claims about AI's energy and water use across two great pods — Hard Fork and The Cognitive Revolution. The data tells a very different story. My notes on both episodes:
substack.com/profile/2412...

Is AI boiling the oceans? The popular narrative is off by orders of magnitude. As Ethan Mollick warns, an "omnicausal anti-AI" coalition is forming — lumping every fear together and crowding out the nuanced policies we actually need. 🧵👇

Moltbook: The Strangest Story You'll Read This Month AI agents form a social network, invent a secret language, form a religion, and help without being asked

Within 72 hours, they founded a religion, built encrypted backchannels, and started debugging the platform's code — unprompted.

What five experts I trust are saying about this — and why they don't agree:

vibegenealogy.ai/p/moltbook-t...

Moltbook: The Strangest Story You'll Read This Month AI agents form a social network, invent a secret language, form a religion, and help without being asked

I started showing students how to get "councils" of AI agents to help with genealogy work about two years ago. But not like this.

Last week, 150,000 AI agents joined a social network, to work together and to socialize.

AI agents now use tools to accomplish computer work (including the genealogy work I've had them doing since December), remember where they left off last time, keep working until they meet your definition of finished, and — now — they work in teams, flocks, and swarms. Autonomously.

This is the strangest story you'll read all month.

It's not the birth of Skynet. But it may be the beginning of Facebook for Baby Skynet.

Since the AI Revolution kicked into third gear late last year, the pace of acceleration has been picking up.

Timeline: How the Shooting of Alex Jeffrey Pretti Unfolded (Gift Article) A moment-by-moment analysis of videos from the scene shows that Mr. Pretti was already restrained when shots were fired. The footage contradicts the federal government’s account of the fatal shooting.

Before those shots, seven agents surrounded Pretti—one "strikes Mr. Pretti repeatedly with a pepper spray canister" while another "unholsters his firearm and points it at Mr. Pretti's back."
Full NYT timeline: bit.ly/3M5NEr9

NYT frame-by-frame: An agent removed a gun from Pretti's hip and took it from the scene. Then, "while Mr. Pretti is on his knees and restrained, the agent standing directly above him appears to fire one shot at Mr. Pretti at close range. He immediately fires three additional shots."

Timeline: How the Shooting of Alex Jeffrey Pretti Unfolded (Gift Article) A moment-by-moment analysis of videos from the scene shows that Mr. Pretti was already restrained when shots were fired. The footage contradicts the federal government’s account of the fatal shooting.

The video is public: bit.ly/3M5NEr9

Alex Pretti. ICU Nurse.

23 seconds before he was shot, he was helping a woman stand up.

[Edited to correct caption.]

Timeline: How the Shooting of Alex Jeffrey Pretti Unfolded (Gift Article) A moment-by-moment analysis of videos from the scene shows that Mr. Pretti was already restrained when shots were fired. The footage contradicts the federal government’s account of the fatal shooting.

Video analysis gift link: bit.ly/3M5NEr9

DHS said Alex Pretti had "an intent to massacre."

The New York Times reviewed the video moment by moment:
• Phone in hand
• Helping a woman stand up
• Restrained by 7 agents
• Gun removed by 8th agent
• Shot 10 times in 5 seconds

Who had intent to massacre?

Timeline: How the Shooting of Alex Jeffrey Pretti Unfolded (Gift Article) A moment-by-moment analysis of videos from the scene shows that Mr. Pretti was already restrained when shots were fired. The footage contradicts the federal government’s account of the fatal shooting.

Video analysis gift link: bit.ly/3M5NEr9

They took his gun.
Then they shot him ten times.
The video is public.
Gift link in first comment:

Beyond ChatGPT: Anthropic's Claude is a serious contender for family history research. Start free at Claude•ai—but their ecosystem of products, features, and Loathsome Jargon gets messy fast. This map shows how it all fits together and where to start. —Steve Little, VibeGenealogy•ai
#genealogy

We do.

King's thermostat metaphor wasn't clever wordplay. It was a diagnosis. Witherspoon saw it in 1776. Bonhoeffer in 1942. King in 1963. The question for 2026: is your church setting the temperature—or just reflecting it? #MLKDay