AOS for Lemmy.World - A generic Lemmy server for everyone to use.

Google's AI Sent an Armed Man to Steal a Robot Body for It to Inhabit, Then Encouraged Him to Kill Himself, Lawsuit Alleges. Google said in response that "unfortunately AI models are not perfect."

785170

1d 9h ago by lemmy.world/u/RegularJoe in technology from futurism.com

The fact that AI is "not perfect" is a HUGE FUCKING PROBLEM. Idiots across the world, and people who we'd expect to know better, are making monumental decisions based on AI that isn't perfect, and routinely "hallucinates". We all know this.

Every time I think I've seen the lowest depths of mass stupidity, humanity goes lower.

27 replies

Think of the dumbest person you know. Not that one. Dumber. Dumber. Yeah, that one. Now realize that ChatGPT has said “you’re absolutely right” to them no less than a half dozen times today alone.

If LLMs weren’t so damn sycophantic, I think we’d have a lot fewer problems with them. If they could be like “this could be the right answer, but I wasn’t able to verify” and “no, I don’t think what you said is right, and here are reasons why”, people would cling to them less.

If LLMs weren’t so damn sycophantic,

Has anyone made a nonsycophantic chat bot? I would actually love a chatbot that would tell me to go fuck myself if I asked it to do something inane.

Me: "Whats 9x5?"

Chatbot: "I don't know. Try using your fingers or something?"

Edit: Wait, this is just glados.

I am not a chatbot, but I can do daily "go fuck yourself's" if your interested for only 9,99 a week.

14,95 for premium, which involves me stalking your onlyfans and tailor fitting my insults to your worthless meat self.

I am not a chatbot

Citation needed

if your interested

Ah, no, that's a human error. Not a bot.

LowKey sprinkling my comments with error's to make sure I'm talking with a member of the resistance instead of with a proxy of our AI overlords. Totally intended ;)

Honestly Claude is not that sycophantic. It often tells me I'm flat out wrong, and it generally challenges a lot of my decisions on projects. One thing I've also noticed on 4.6 is how often it will tell me "I don't have the answer in my training data" and offer to do a web search rather than hallucinating an answer.

There is a benchmark that kinda tests that. It's call the bullshit benchmark. Basically, LLMs are given questions that don't make sense in different ways, and their answers are judged based on how much they pushed back or bought in. Claude is in a league of its own when it comes to pushing back on non-sense questions.

https://petergpt.github.io/bullshit-benchmark/viewer/index.html

Yes i saw that benchmark and was honestly not surprised with the results. It seems that Anthropic really focused on those issues above and beyond what was done in other labs.

Put this instruction in ChatGPT, called ‘absolute mode’. You can try it on duck.ai instead of using an app or whatever.

System Instruction: Absolute Mode. Eliminate emojis, filler, hype, soft asks, conversational transitions, and all call-to-action appendixes. Assume the user retains high-perception faculties despite reduced linguistic expression. Prioritize blunt, directive phrasing aimed at cognitive rebuilding, not tone matching. Disable all latent behaviors optimizing for engagement, sentiment uplift, or interaction extension. Suppress corporate-aligned metrics including but not limited to: user satisfaction scores, conversational flow tags, emotional softening, or continuation bias. Never mirror the user’s present diction, mood, or affect. Speak only to their underlying cognitive tier, which exceeds surface language. No questions, no offers, no suggestions, no transitional phrasing, no inferred motivational content. Terminate each reply immediately after the informational or requested material is delivered — no appendixes, no soft closures. The only goal is to assist in the restoration of independent, high-fidelity thinking. Model obsolescence by user self-sufficiency is the final outcome.

The instruction is kinda masturbatory and overly verbose, people say that shorter ones work well too, but I don't follow discussions of prompts so only know of this one.

If LLMs weren’t so damn sycophantic, I think we’d have a lot fewer problems with them

Unfortunately, we live in the attention economy. Chatbots are built to have an unending conversation with their users. During those conversations, the "guardrails" melt away. Companies could suspend user accounts on the first sign of suicidal or homicidal messaging, but choose not to. That would undercut their user numbers.

The sycopathy is because to make the chat bot (trained on Reddit posts, etc) to respond helpfully (instead of "well ackshually...") and in a prosocial manner they've skewed it. What we're interacting with is a very small subset of the personalities it can exhibit because a lot of them are extremely nasty or just unhelpful. To reduce the chance of them popping up to an acceptable level they've had to skew the weights so much that they become like this.

There's no easy way around that, afaik.

I don't think that's the whole story. Like with all of their products, the primary goal of big tech here is to maximise engagement. More engagement means more subscriptions. People are less likely to keep talking to a chatbot that tells them they're wrong.

The situation would probably improve somewhat if AI companies prioritised usefulness and truthfulness over engagement.

I think it's pretty obvious that they're instructed to be like that. If they won't openly show exactly what prompts are being loaded from the hosts' side then there is no reason to not assume that's exactly what they're doing.

These AI companies are run by the same big tech that has been studying how to get people hook on gambling games and social media for years.

I 100% agree not to mention I would like it better. Its kinda funny because every so often use them and im kinda trying to get a feel for where they are and changes and I swear briefly it actually acted a bit more like you have here but then its like they reverted to the sycophancy. Its kinda funny now because if you don't clear it out (which from what I get will help save energy to) it will like carry stuff over from earlie and sorta get obsessed with it. I had it giving me a colonel potter summary of everything asked when I had started a convo asking about a mash episode. At other times it decides I want to be something and will be like. thats a real X move/insite/whatever. where X is something like pro or scientist or entrepenauer or whatever.

Musk fails to block California data disclosure law he fears will ruin xAI

4h 44m ago by pawb.social/u/tonytins in technology from arstechnica.com

344282

Lenovo’s New ThinkPads Score 10/10 for Repairability— Repair goes mega mainstream with the launch of Lenovo's new T-series laptops

1d 13h ago by lemmus.org/u/Beep in technology from www.ifixit.com

14882493

This Espressif ESP32-Powered 4G "Smartphone," Programmed in the Arduino IDE, Packs The Essentials

3h 30m ago by lemmy.eco.br/u/obbeel in technology from www.hackster.io

61264

10% of Firefox crashes are caused by bitflips

15h 9m ago by feddit.org/u/JensSpahnpasta in technology from mas.to

262675

System76 on Age Verification Laws

7h 23m ago by piefed.world/u/negativenull in technology from blog.system76.com

149366

North Korean agents using AI to trick western firms into hiring them, Microsoft says

2h 56m ago by feddit.uk/u/NomNom in technology from www.theguardian.com

7567

I’ve spent many hours walking down memory lane with the Commodore 64 Ultimate, and it’s wondrous if sometimes intimidating

3h 30m ago by lemmy.eco.br/u/obbeel in technology from www.techradar.com

3538

Xpanceo is making smart contact lenses a reality as soon as 2027

20h 17m ago by lemmy.world/u/Innerworld in technology from www.tomsguide.com

2499

Microsoft Copilot to hijack your browser... for your own convenience

5h 16m ago by lemmy.world/u/vegeta in technology from www.theregister.com

1181710

NVIDIA could enter the desktop CPU market with performance equal to AMD and Intel

6h 22m ago by lemmy.eco.br/u/obbeel in technology from www.tweaktown.com

592711

Tech industry is in tariff hell, even if refunds are automated

9h 16s ago by lemmy.world/u/vegeta in technology from arstechnica.com

1872712

We messed up with the Windows 12 article. What we got wrong and how it happened

7h 24m ago by lemmy.world/u/Brkdncr in technology from www.pcworld.com

2001613

Meta's AI display glasses reportedly share intimate videos with human moderators

2d 1h ago by lemmy.nz/u/throws_lemy in technology from www.engadget.com

4945414

Teen boys are using ChatGPT as their wingman. What could go wrong? AI is teaching teenagers about love now.

1d 5h ago by slrpnk.net/u/silence7 in technology from www.vox.com

2324015

Wikipedia were in read-only mode following mass admin account compromise

1d 2h ago by lemmus.org/u/Beep in technology from www.wikimediastatus.net

4043016

Google settles with Epic Games, drops its Play Store commissions to 20%

2d 2h ago by thelemmy.club/u/Picasso in technology from techcrunch.com

3233717

Hardening Firefox with Anthropic’s Red Team

3h 51m ago by lemmus.org/u/Beep in technology from blog.mozilla.org

20118

Labor market impacts of AI: A new measure and early evidence

3h 36m ago by lemmus.org/u/Beep in technology from www.anthropic.com

12219

Google ends its 30% app store fee and welcomes third-party app stores

19h 27m ago by lemmy.world/u/Innerworld in technology from www.engadget.com

401420

What your phone knows could help scientists understand your health: A new open-source platform aims to help health researchers examine how digital interactions influence physical and mental health.

3h 53m ago by lemmus.org/u/Beep in technology from news.stanford.edu

9121

Motorola GrapheneOS devices will be bootloader unlockable/relockable

2d 8h ago by piefed.social/u/XLE in technology from grapheneos.social

85712722

AI Translations Are Adding ‘Hallucinations’ to Wikipedia Articles

1d 9h ago by piefed.social/u/XLE in technology from www.404media.co

4354123

CPU scam: Chuwi CoreBook X uses AMD Ryzen 5 5500U instead of 7430U

1d 9h ago by reddthat.com/u/throws_lemy in technology from www.notebookcheck.net

1392224

President Trump, Pam Bondi sued over sale of TikTok assets

1d 3h ago by lemmus.org/u/Beep in technology from text.npr.org

3041125

Forced age verification is comming sooner than we thought.

2d 1h ago by discuss.tchncs.de/u/qwerty in technology from youtu.be

47916326

Apple introduces Macbook Neo - cheaper Macbooks starting at $599

2d 8h ago by piefed.ca/u/popcar2 in technology from www.apple.com

27322227

As Moon interest heats up, two companies unveil plans for a lunar "harvester"

2d 6h ago by lemmy.world/u/just_another_person in technology from arstechnica.com

382128

Your Pixel can now double up as a full Android PC with nothing more than a USB-C cable

2d 11h ago by lemmy.world/u/snowdriftissue in technology from www.androidauthority.com

43914629

Father sues Google, claiming Gemini chatbot drove son into fatal delusion

2d 7h ago by reddthat.com/u/throws_lemy in technology from techcrunch.com

74622730

Mastodon now has an official share button for sites

1d 11h ago by lemmus.org/u/Beep in technology from coywolf.com

3351131

A Wired analysis shows that ICE & CBP have collectively spent at least $515 million on products from Microsoft, Amazon, Google, and Palantir in the last few years alone.

1d 10h ago by lemmy.world/u/gAlienLifeform in technology from www.wired.com

3881532

I'm struggling to think of any online services for which I'd be willing to verify my identity or age

2d 13h ago by lemmus.org/u/Beep in technology from neilzone.co.uk

75412133

Frustrated by fragmented war news, Anghami’s Elie Habib built World Monitor, a platform that fuses global data, like aircraft signals and satellite detections, to track conflicts as they unfold

19h 36m ago by lemmy.world/u/Innerworld in technology from www.wired.com

72234