Old Fucking Punk @lwriemen@librem.one

I saw a thread today explaining why ChatGPT is great for making regular programmers into 10x programmers. Most of the thread was misguided, including the part where 10x apparently refers to *output*, not quality.

Anyway, I threw a few code snippets at ChatGPT, asking it to explain them, find bugs, or improve them. It did pretty well with the first 5-ish queries but then went wildly off the rails when I gave it a one-liner of Kotlin that just did some multiplication.

Confidently, and self-assuredly, it explained that this was likely vulnerable to some inputs being zero. (It missed the integer overflow bug, and in fact specifically claimed that the code prevented overflow.) When asked to improve the code, it started adding checks to short-circuit on 0 (unnecessary for logic, but could improve performance I guess?), negate the output if there were an odd number of negative numbers (wrong), and coerce floating-point numbers to integers (but they were already integers!)

And of course it just doubles down on mistakes, even when prompted to reconsider.

This is going to cause so, so, so many bugs in the hands of junior developers. And awkward conversations in code review, too...

An "openly racist" teacher in Florida is practicing what the governor's dog whistles preach. https://popular.info/p/florida-english-teacher-pushing-book?publication_id=1664&post_id=95514667&isFreemail=false

We've learned this lesson before: Absent strict protections, any sensors that collect data and footage about people will eventually become a tool of police surveillance--even self-driving cars. https://www.vice.com/en/article/v7dw8x/san-francisco-police-are-using-driverless-cars-as-mobile-surveillance-cameras

Where are the voice actors of the fediverse? I miss my VA friends from the birdsite. I do animation and motion graphics, so I not only really appreciate the work, but I also sometimes like to collaborate with people I know :)

#VoiceActor #VoiceArtist #Narrator #Actor #Acting

The status quo, where many (most?) people have a primary digital identity (their email) which carries a corporate brand. Once they're established (often without any forethought), people are loath to change them. They then have to put up with whatever crap that corporate dishes out.

We need an easy way for people (e.g. families) to manage their own identity and communications, like email addresses. I think everyone should have their own domain name & email that uses that domain name.

I should have phrased that slightly differently: The question isn't why -- in a late-state capitalist nation like ours -- sleazy, security-indifferent companies like Experian exist. It's why our corrupt system of government permits them to do what they do.

ICYMI, today's pro tip, the need for which is still pissing me off:

Adobe has an art-sucking AI training on every file you create. Here's how to turn it off: https://www.howtogeek.com/858952/adobe-is-using-your-data-to-train-ai-how-to-turn-it-off/

#aiart #ai #fighttherobots

Why does "credit reporting" Experian still exist? This rogue company and industry have demonstrated again and again that they cannot be trusted with the detailed personal and financial information it holds about you and almost everyone else in the U.S. The latest: https://krebsonsecurity.com/2023/01/identity-thieves-bypassed-experian-security-to-view-credit-reports/

This story -- "US farmers win right to repair John Deere tractors" -- is bizarre. It takes for granted that corporate control-freakery requires special exceptions for "buyers" of products they don't truly own -- and this deal maintains Deere's control in major ways. And the story ignores some key history, namely that Deere promised buyer freedom-to-repair years ago, but reneged. https://www.bbc.com/news/business-64206913

Global pollinator losses causing 500,000 early deaths a year – study - https://www.theguardian.com/environment/2023/jan/09/global-pollinator-losses-causing-500000-early-deaths-a-year-study a bold claim, but we certainly need to protect bees et al.

The Old Reliable Paper of Record is at it again. Here is the latest in normalizing extremists and establishing a bogus equivalence between the two major parties, ONE of which is working feverishly to wreck things and ultimately bring down democracy.

The New York Times political coverage cannot be saved. It can only be torn down and reconstruction.

@joshuafoust
What I would like to read:
"We put tech billionaires in a plane controlled by experimental tech.
You won't believe what happened next."

So basically: like the January 6th thing in the US, this is just a lot of meaningless noise. But remember: the US is never in danger of suffering a coup because it's the only country in the world without a US embassy. Whereas Brazil........

What's happening in Brazil right now: weird bolsonarista fascists are breaking into federal buildings and making a mess. It's not a serious coup attempt in the sense that these are just random civilians and the whole thing is a circus. What IS concerning is that the DF (Federal District) police are letting them do whatever they want, which is why Lula decreed a federal intervention in the DF.

Single-use #plastic cutlery and plates to be banned in England - https://www.theguardian.com/environment/2023/jan/08/single-use-plastic-cutlery-and-plates-to-be-banned-in-england no perfect, but a good move by UK gov for a change...

So the people at Koko— an online mental health intervention laminate that licenses itself out to other platforms to make use of (Tumblr, facebook, etc.) have apparently just admitted that they tested gpt-3 assisted mental health interventions on some 4,000 people in mental health crisis, without said people's informed consent.

See here: http://web.archive.org/web/20230107210632/https://twitter.com/RobertRMorris/status/1611450197707464706

And followup "Clarification" thread here: http://web.archive.org/web/20230107191117/https://twitter.com/RobertRMorris/status/1611775514590740480

Now, at this point, I really shouldn't have to explain to people why this is… very very bad on a number of levels, but to those who maybe still don't get it, and to anyone ever even THINKING of doing anything like this, PLEASE read the "Digital Futures in Mind" Report:

https://automatingmentalhealth.cc/recommendations

This report was funded by the @mozilla foundation, and coordinated by Piers Gooding, with further authorship by Jonah Bossewitch, Lydia X. Z. Brown, Leah Harris, James Horton, Simon Katterl, Keris Myrick, Kelechi Ubozoh, and Alberto Vásquez Encalada (and by way of full disclosure, i wrote the foreword).

It covers the promises, pitfalls, and outright depredations of the Existing interventions in the "AI" assisted mental health space, while also examining and interrogating the proposed use cases and laying bare their implications.

There are several things in the references you should take in, too, but if nothing else PLEASE take the time to read the report itself.

This is a mess that didn't need to happen, and definitely doesn't need to be repeated.