GitHub is Leaking Trump's Plans to 'Accelerate' AI Across Government

[highplainsdem]

Source: 404 Media

The federal government is working on a website and API called “ai.gov” to “accelerate government innovation with AI” that is supposed to launch on July 4 and will include an analytics feature that shows how much a specific government team is using AI, according to an early version of the website and code posted by the General Services Administration on Github.

The page is being created by the GSA’s Technology Transformation Services, which is being run by former Tesla engineer Thomas Shedd. Shedd previously told employees that he hopes to AI-ify much of the government. AI.gov appears to be an early step toward pushing AI tools into agencies across the government, code published on Github shows.

“Accelerate government innovation with AI,” an early version of the website, which is linked to from the GSA TTS Github, reads. “Three powerful AI tools. One integrated platform.” The early version of the page suggests that its API will integrate with OpenAI, Google, and Anthropic products. But code for the API shows they are also working on integrating with Amazon Web Services’ Bedrock and Meta’s LLaMA. The page suggests it will also have an AI-powered chatbot, though it doesn’t explain what it will do.

The Github says “launch date - July 4.” Currently, AI.gov redirects to whitehouse.gov. The demo website is linked to from Github (archive here) and is hosted on cloud.gov on what appears to be a staging environment. The text on the page does not show up on other websites, suggesting that it is not generic placeholder text.

-snip-

Read more: https://www.404media.co/github-is-leaking-trumps-plans-to-accelerate-ai-across-government/

[Lonestarblue]

1. We recently saw RFK's AI-generated report filled with false information and references to nonexistent research.

The NYT also recently published info about an AI-generated book list. Most of the books and authors did not exist. Given the corruption and deliberate promotion of propaganda and lies by this administration, any AI- generated reports and data should be treated as pure fiction.

[highplainsdem]

2. Nothing from genAI should be trusted until it's checked. Of course we can't trust this regime of thugs and liars anyway.

[Harker]

3. Crap.

[highplainsdem]

9. Exactly what this will be.

[Harker]

12. If I want to have my questions questioned, or to be subjected to vague, repetitive drivel posing as useful information

I'll go straight to AI.

[louis-t]

4. Apparently, these thugs are not confident they can screw things up enough by themselves.

My experience with AI so far has been photos of people with 6 fingers on each hand and a guy that sent me setlists with the keys of the songs listed. Unfortunately, most of the keys were wrong.

[highplainsdem]

10. Sounds typical of AI.

[JHB]

5. AI can be a useful tool if you're deliberate about what you're doing. Unfortunately...

...the only "deliberate" thing going on here is the pell-mell shoving of these inadequately-tested systems into active use to serve some grand fantasy about replacing "obsolete" government workers (and, of course, because the owners of these systems are huffing the scent of money in their pockets).

[highplainsdem]

6. It's an unnecessary tool that dumbs down and deskills users, and makes lots of mistakes that are

hard to catch. Pretty much impossibly to catch if, for instance, you're using it to summarize a lot of data or writing. I posted the other day about UK plans to use generative AI to summarize public feedback on proposed policies. The Scottish Parliament did one genAI trial checked against human work, and though there were differences in the results, they decided the differences weren't important enough to worry about. But with genAI, results and error rates can.vary widely from one use to another, and they can't count on one not-too-bad accuracy rate to be typical.

[Bayard]

7. This has disaster written all over it

[highplainsdem]

11. Yep.