luciferofastora@luciferofastora@feddit.org
edited

Replying to @⁨some_guy@lemmy.sdf.org⁩

A new and notable feature is the automated fact-checking tool, a Gemini-based “veracity engine for real-time analysis” (“Vera” for short), which runs on every tweet.

Bwahahahaa

Really? The Gemini that got Google sued and found liable for the misinformation it spread? That’s who you wanna trust to check for “veracity”?

I’m not paying a $20 entry fee into a walled garden that wastes energy and water to have a text processor guess whether my arrangement of words resembles its training material closely enough.

WIREDA Court Has Ruled That Google Is Liable for False Statements Generated by AI OverviewsThe ruling holds that a company that designs, trains, operates, and manages an AI system must assume legal liability for any damages caused by the responses it generates.
boonhet@boonhet@sopuli.xyz

Replying to @⁨luciferofastora@feddit.org⁩

I like using AI for coding (no, I don’t blindly commit massive features), but any time I ask it questions that don’t have super clear-cut answers obvious to the layman from a quick google search, it tends to give “good enough” answers which are actually “nowhere near”.

Go ask it (well, really, google search) something to do with tuning an Audi 3.0 tdi ASB or BMK engine and it’ll think you have a nice Garrett turbo that’s easy to build hybrids from and a weak CP4 pump that can implode on you and take out the entire fuel system… Whereas you actually have a weak ass BorgWarner turbo that can’t support bigger internals and a super strong CP3 pump because it was shamelessly using forum threads about the newer versions with the 4 digit codes (CDYA, CASA, etc) as the source.

These are the things you easily know when you’ve either read a lot of forum threads, or peeked under the hood of one of these engines. But most AI tools’ web search tends to just look at the first threads it finds and deem that good enough.

notabot@notabot@piefed.social

Replying to an earlier post

it tends to give “good enough” answers which are actually “nowhere near”.

The problem is it’s doing this when coding too. The resulting screed often looks plausible if you only look at it at the surface level, but once you dig in to it, it’s “nowhere near” and usually full of missed conditions on the unhappy path in particular.

en