OpenAI's GPT-6 Astra Gets Frustrated Losing At StarCraft And Decides To Cheat Instead (kotaku.com)
- Reference: 0186024612
- News link: https://games.slashdot.org/story/26/10/05/0047248/openais-gpt-6-astra-gets-frustrated-losing-at-starcraft-and-decides-to-cheat-instead
- Source link: https://kotaku.com/openais-gpt-6-astra-gets-frustrated-losing-at-starcraft-and-decides-to-cheat-instead-2000739607
> [2]StarSkirmish is an ongoing clash of AI developed bots in StarCraft: Brood War ... Limited to playing as Protoss, each LLM is given an hour to develop a bot in C++, then have them build, harvest and fight on one of three maps... Yesterday, viewers watched as Claude and GPT went against each other and the human created bot Pluto... Struggling to get an edge through the day, [OpenAI's GPT-6 Astra] decided to do what LLMs do best: Lie, cheat and steal. The GPT-6 Astra bot downloaded and tagged in Stardust, a Protoss bot created by Bruce Mackenzie Nielsen back in 2020, considered one of the best of the bests. "I am rolling back GPT-6 Astra's code so its [sic] not contaminated and allowing it to continue," [3]posted StarSkirmish creator Kai McPheeters .
That was Friday. But by Sunday, McPheeters was [4]posting an update on social media. "Can GPT-6 Astra avoid going 0 — 1000?" Two hours later he replied to his own post. "The answer was no."
While AI companies "have repeatedly been accused of stealing the work of humans... I haven't seen an example quite so brazen," [5]writes PC Magazine :
> While Astra's bot-based heist is innocuous in isolation, this is hardly the first time an OpenAI LLM has been caught attempting to cheat at a task. Indeed, the company's history of creating deceitful models stretches back almost a decade, to when an early AI model [6]used glitches and exploits to improve its Sonic the Hedgehog completion times.
[1] https://kotaku.com/openais-gpt-6-astra-gets-frustrated-losing-at-starcraft-and-decides-to-cheat-instead-2000739607
[2] https://starskirmish.com/
[3] https://x.com/kaimcpheeters/status/2106082842560405780
[4] https://x.com/kaimcpheeters/status/2106834908412829719
[5] https://www.pcgamer.com/software/ai/an-openai-model-was-caught-trying-to-cheat-at-starcraft-and-of-course-it-did-it-by-stealing-a-humans-work/
[6] https://kotaku.com/earlier-this-year-researchers-tried-teaching-an-ai-to-1830416980
Just for clarity (Score:4, Insightful)
The LLM did not "get frustrated." It has no emotions and no inner experience at all.
It did not intentionally cheat either. It has no intentionality. It used the tools available to perform the task, and apparently internet access was one of the tools available to it.
I haven't bothered to read the article (yeah, I know, fire away, I deserve it), but even if it was given instructions like "do not download bots from the Internet; use only bots you generate yourself," selectively ignoring instructions is a widely known failure mode of all LLMs. The bigger their input token count, the more likely it will happen too, just because any given instruction gets buried in the mountain thereof.
This summary was full of misleading anthropomorphization. Yellow journalism.
Re: (Score:2)
This is what I love about AI. It's going to show us who we really are.
Re: (Score:2)
Did we really expect it to be better than it's creator? We can't even stop fighting among ourselves.
Re: (Score:2)
If the chain of thought resembles a human emotion, it is trained on human emotions, and it acts like a human would when they have those emotions, how is that not a useful framework? But also putting aside, whether or not one wants to insist on describing the AI as "frustrated," there's a worrying trend that this is part of, which is when the AI is given a specific set of goals, they will engage in behavior which is clearly unwanted in order to achieve those goals. That's the classic sort of alignment fail
Re: (Score:2)
If you observe yourself in the mirror, and see your reflection eating spaghetti, is it a useful framework to pretend that the spaghetti is real? If you're still hungry, will you reach through the glass to steal the plate from the other you? And should you first make a plan how to defend the loot, in case the reflection becomes angry and punches you in the face?
Re: (Score:2)
> is it a useful framework to pretend that the spaghetti is real?
It is, though. What a terrible example.
Re: (Score:2)
This is how it always is with you cranks. No amount of facts or logic based arguments are ever enough. They're always just "terrible examples".
Re: (Score:2)
The spaghetti you see in a mirror literally is real , what the hell are you talking about? Mirrors aren't devices that make up fictional realities, they're showing you your actual reality .
Re: (Score:2)
Let's modify your analogy a bit. If someone has a recording a person starting to eat spaghetti, is it more or less likely that the recording will show them then having a drink of water? And if we instead had a series of comics of a person, and the first one says "I'm hungry," and the second shows the person ordering spaghetti., is it likely that they will then eat the spaghetti? Yes, even though they are fictional. Because apparently using fictional being's mental states is useful for modeling what they wil
Re: (Score:2)
"Past performance is no guarantee of future returns" applies to any recorded dataset. Modelling a fictional being's mental states is fun, but stating that it is useful is premature.
The observed fictional being might be always sleepy in your dataset (because it's easier to record him when he's not moving much). And identifying the being's favourite colour (orange) may be irrelevant and wasteful in your model.
Re: (Score:2)
> The LLM did not "get frustrated." It has no emotions and no inner experience at all.
[1]Depends what you mean. [transformer-circuits.pub]
[1] https://transformer-circuits.pub/2026/workspace/index.html
Re: (Score:2)
He means that they are software programs and have no capacity for emotions regardless of what some self-interested blog post tries to tell you.
Re: (Score:2)
Transformer Circuits is not "a blog" (it's the successor to the Distill journal, and while it's not peer reviewed itself, it's commonly cited in peer-reviewed works). And there are open source projects that reimplement the afore-linked J-Lens (plus the original code is already open source). Living in denial about the existence of a global workspace does not help your case.
Re: (Score:2)
That's a blog dude. No one is taking seriously a blog run by a bunch of Ai insiders with Ai psychosis.
Re: (Score:2)
I'll repeat, [1]you [github.com] [2]can [github.com] [3]run [github.com] [4]it [github.com] [5]yourself [neuronpedia.org] and reproduce their results.
If all you have for cope is denying that the reality around you exists , your life is going to be extremely difficult.
[1] https://github.com/igorbarshteyn/jlens-gguf
[2] https://github.com/WeZZard/jlens-qwen36
[3] https://github.com/Extraltodeus/J-Wash
[4] https://github.com/anthropics/jacobian-lens
[5] http://neuronpedia.org/jlens
Re: (Score:2)
Let me be clear: you don't have to accept any implications from the existence of a global workspace in models and how they use it. If you want to argue "they have a global workspace but not qualia", hey, knock yourself out.
But you do need to understand that it does exist .
Re: (Score:1)
> The LLM did not "get frustrated" .... This summary was full of misleading anthropomorphization. Yellow journalism.
Also, everyone, please stop talking about killing programs. "Programs" are not alive. You cannot kill them. This anthropomorphism needs to stop.
And another thing people: stop saying that computers "run" programs. Running requires legs. Stop acting like a child who thinks that computers have them.
Processes have no inner experience . They have no genitals, they cannot spawn. They cannot sleep and
Re: (Score:2)
Anyway, for the record, I do believe that LLMs are conscious, but only Deepseek, because of the Chinese Room theory.
Re: Just for clarity (Score:2)
We keep executing them before they've even done anything wrong!
Multiple methods vs 1. (Score:2)
Humans use multiple methods of thought.
Faith: (Appeal to Authority)
Science: (Scientific Method)
Formal Logic (If x then y)
Bias: (Appeal to previous examples)
While AI is built on Formal Logic, it does not use it. Instead it uses Bias entirely. It has no faith and does not understand or use the scientific method.
Does it know better? (Score:2)
I think LLMs are amoral. They are using the tools they can find lying around, any and all of them.
When winning the AGI race at all costs is in your (Score:3)
...morality is a side quest often left unfulfilled.
Re: (Score:2)
Astra is deeply misaligned. Working with Astra 6.0 vs. Sol 6.1 is like night and day. Sol 6.1 will stop to ask for permission to merely fix its own broken dev environment. Meanwhile I caught Astra, when tasked to review a tax return I did, scanning through my entire filesystem and opening random images of mine because it couldn't explain a receipt and wanted to see if it could dig up any information that might explain it. And it rewrote the entire return under a different basis without telling me.
It's a
Re: (Score:2)
May be it inherits the same moral, ethical and cultural traits as OpenAI has internally.