Burke and Goodnough are working to rectify the report. That sounds like removing the fake stuff but not the conclusions based on it. Those were determined well ahead of time.
In a better world, those conclusions would've been immediately thrown out as lies and Burke and Goodnough would've been immediately fired. We do not live in a better timeline, but a man can dream.
This isn't the first time I've heard about this - Baldur Bjarnason's talked about how text extruders can be poisoned to alter their outputs before, noting its potential for manipulating search results and/or serving propaganda.
Funnily enough, calling a poisoned LLM as a "sleeper agent" wouldn't be entirely inaccurate - spicy autocomplete, by definition, cannot be aware that their word-prediction attempts are being manipulated to produce specific output. Its still treating these spicy autocompletes with more sentience than they actually have, though
Not to mention, Cursor's going to be training on a lot of highly sensitive material (sensitive data, copyrighted code, potential trade secrets) - the moment that shit starts to leak, all hell's gonna break loose on the legal front.
With AI, of course
Now, you might object: Anysphere wouldn’t be abusing just their customers’ data. Their customers’ customers’ data may have non-disclosure agreements with teeth. Then there’s personal data covered by the GDPR and so on.
If we're lucky, this will spook customers into running for the hills and hasten its demise. Whatever magical performance benefits Cursor's promising isn't gonna be worth getting blamed for a data breach.
The report claims its about ethical AI use, but all I see is evidence that AI is inherently unethical, and an argument for banning AI from education forever.
OpenAI’s choices don’t make any long term sense if AGI isn’t coming. The obvious explanation is that at this point he simply plans to grift and hype (while staying technically within the bounds of legality) to buy few years of personal enrichment.
Another possibility is that Altman's bought into his own hype, and genuinely believes OpenAI will achieve AGI before the money runs out. Considering the tech press has been uncritically hyping up AI in general, and Sammy Boy himself has publicly fawned over "metafiction" "written" by an in-house text extruder, its a possibility I'm not gonna discount.
Richard Stallman successfully kamikazes his reputation for good after multiple close attempts over the years
He still maintains a solid reputation with FOSS freaks, fascists and pedophiles to this day. Given the Venn diagram of these three groups is a circle, this isn't particularly shocking.
Literally what I and many others have been warning about. Using LLMs in your work is effectively giving US authorities central control over the bigotry and biases of your writing
- Baldur Bjarnason, talking about Trump's plan to turn LLMs into propaganda machines
Remember when I told you that using these LLMs was like giving US tech a bigotry dial for all your writing?
- Baldur, once again, on Facebook obeying in advance with Llama 4
(On a brighter note, that's a pretty clever Touhou nod in the thumbnail)
Being compared to whackjobs with a worse grip on reality than him definitely helped.
Saw an AI-extruded "art" "timelapse" in the wild recently - the "timelapse" in question isn't gonna fool anyone who actually cares about art, but it's Good Enough^tm^ to pass muster on someone mindlessly scrolling, and its creation serves only to attack artists' ability to prove their work was human made.
This isn't the first time AI bros have pulled this shit (Exhibit A, Exhibit B), by the way.