Core insights:

Bots are bad at saying "I don't know". They rather offer incorrect info. Or speculate.

Bots fabricate links and cite syndicated and copied versions of articles.

Multiple bots seem to bypass bot exclusion protocols.

Content licensing deals provide no guarantee of accurate citation.

Premium bots are worse than their free counterparts.

Bottom line: You can't really trust AI search engines (yet).

(via Columbia Journalism review)

https://www.cjr.org/tow_center/we-compared-eight-ai-search-engines-theyre-all-bad-at-citing-news.php

Months after people started talking about it on the Fediverse, my Google has *finally* started giving me "AI Overviews" when I ask it questions.

I've been using it for the past week, and I must be honest, I'm pretty well impressed. I have yet to see one hallucination, nor have I been advised to throw myself off a bridge or drink detergent to cure my eczema, or any of the other crazy things I've seen in screenshots.

I'm sure those things exist, as with any LLM, but I also get a feeling that sometimes, the anti-AI crowd goes out of their way to catch these LLMs out by asking questions that are the most likely to result in these laughable situations, perhaps multiple times in different ways until they get one, just so they can post a screenshot and say, "Lookie AI sucks waaaaah!"

(Of course, when I say it's pretty good, I do mean it's pretty good from a purely consumer perspective. Obviously it reduces the traffic to the sites that serve as sources of that information, so those creators lose out. That's a separate discussion. )

In Changping, a Xiaomi "dark factory" runs 24/7 with no need for humans. Just AI and robots doing work completely autonomously. The factory is a blueprint for the future of global manufacturing.

https://www.news.com.au/finance/business/manufacturing/chinese-companys-dark-factory-will-no-human-workers-soon-be-the-norm/news-story/9468c5bc380108deba4e55a95d6c28d4

Another win

#AI https://flipboard.com/@fastcompany/fast-company-l36v8lfpz/-/a-z6EK4gWSSNOdKeWuLjv0YA%3Aa%3A3199563-%2F0

The DC Circuit's decision in #ThalerVPerlmutter is out, ruling that #copyright #law requires that #authors be #human, not #AI, reasoning to this conclusion from the structure of the copyright statute (e.g., duration is based on the lifetime of the author)

https://media.cadc.uscourts.gov/opinions/docs/2025/03/23-5233.pdf

via NYU Law’s Jeanne Fromer

A Columbia Journalism Review study found that AI search engines frequently cite incorrect sources, with an average error rate of over 60%.

Paid versions actually performed worse than free versions.

In the meantime, countless people place their trust in the confident way they present the answers, of what is, most of times, complete nonsense.

https://arstechnica.com/ai/2025/03/ai-search-engines-give-incorrect-answers-at-an-alarming-60-rate-study-says/

“By firing skilled human workers & replacing them w/spicy autocomplete, Musk is assuming his final form as both…kind of boss who can be conned into replacing you w/a defective chatbot *&* as…fast-talking sales rep who cons your boss. Musk is transforming key government functions into high-speed error-generating machines whose human minders are only the payroll to take the fall for the coming tsunami of robot fuckups.”

https://pluralistic.net/2025/03/18/asbestos-in-the-walls/

#Trump #FTC Removes Posts Critical of #Amazon, #Microsoft, & #AI Companies

Business-guidance content published during the #Biden admin has been removed from the Federal Trade Commission website.

The Trump admin’s FTC removed 4yrs worth of business guidance blogs as of Tues morning, including important #ConsumerProtection info related to AI & the agency’s landmark #privacy lawsuits under fmr chair Lina Khan against companies like Amazon & Microsoft.

#law #broligarchy
https://www.wired.com/story/federal-trade-commission-removed-blogs-critical-of-ai-amazon-microsoft/