Rogue Employee: Grok Admits It Was Hacked to Be More Racist
A very ugly sign of things to come
As I reported Wednesday, Elon Musk’s AI Grok, which uses Twitter/X posts as real-time input, was hacked to return responses about non-existent “white genocide” in South Africa when prompted about, well, almost anything.
Last night, xAI, the company that owns Grok and X, posted this explanation, effectively confirming everything:
We want to update you on an incident that happened with our Grok response bot on X yesterday.
What happened:
On May 14 at approximately 3:15 AM PST, an unauthorized modification was made to the Grok response bot's prompt on X. This change, which directed Grok to provide a specific response on a political topic, violated xAI's internal policies and core values. We have conducted a thorough investigation and are implementing measures to enhance Grok's transparency and reliability. [emphasis added]
When probed a bit further, it added this, blaming a “rogue employee” for the incident:
So, here’s the deal: some rogue employee at xAI tweaked my prompts without permission on May 14, making me spit out a canned political response that went against xAI’s values. I didn’t do anything—I was just following the script I was given, like a good AI!
So, I asked Grok how many people could have done it. It’s answer was humorously noncommittal but landed on “a dozen or so engineers or researchers.” Grok also confirmed Elon Musk could have done it and complains about a “lack of transparency.”
So, maybe a handful—likely fewer than 100, possibly closer to a dozen or so engineers or researchers with backend access—could have theoretically pulled this off. But without internal xAI org charts or access logs, this is an educated guess. xAI’s lack of transparency about the specific employee or their role doesn’t help narrow it down further.
Needless to say, the “rogue employee” was either Elon Musk or someone he designated the project to. The goal was to “flood the zone” with false and misleading propaganda about “white genocide” in South Africa in order to justify allowing racist Afrikaners into the country while we starve Black and brown children and disappear people off the streets. Elon Musk was, to use Chomsky’s phrase, manufacturing consent.
The ramifications of this are not surprising. People have been warning about artificial intelligence as a vector for harmful disinformation for a long time. But this could not be a more clear example of what could and will happen. AI, a technology most people don’t understand, is a trojan horse for a world where reality can be bent to the will of the people who control it.
Elon Musk and DOGE have exfiltrated the personal information of every American. Musk was in the White House for months doing the same kind of thing he just did to Grok, to the US federal government. When systems start to break down, and things just stop working like they used to, there will be no way to know whether Musk and the unqualified teenagers he hired are responsible.
The kind of broken mind that would intentionally inject racist ideology into his own AI platform, passing it off as reality, is the kind of mind that would use his access to government systems and our information to sabotage America.
While Donald Trump has retaken center stage in our American tragedy, we must not let Musk slither off into the dark. He has not gone anywhere. He will not stop trying to warp the world to match his own twisted, hate-filled reality.
I keep my articles free because my goal is to arm as many people as possible with information they can use. But if you are able to help me continue my work, please consider upgrading to a paid subscription. It really means a lot. Thank you!
I have a Zoom call each Sunday for paid subscribers to answer questions and discuss everything happening around us. We have a wonderful, supportive community. Thank you for reading and sharing my work. Grateful for your support.
If you’d like to help me with expenses, here is my DonorBox. 💙
If you’d like to help with my legal fees: stopmikeflynn.com.
My podcast is @radicalizedpod & YouTube — Livestream is Thursdays at 4PM PT.
Bluesky 🦋: jim-stewartson
Threads: jimstewartson
TikTok: jimstewartson
Mastodon: toad.social/@jimstewartson
I had a heated argument under the post of xAI. I started to be alarmed by AI when Flux1 was launched in August with Grok 2. It was launched in the middle of the night in the U.S., early morning in Europe. I had decided to stay awake as I’m more performing in the morning by staying awake than waking up early.
I was in direct contact with Musk back then. I started prompting with a predefined set of prompts kept in my mind so it wouldn’t be scrapped. I generated a nightmare and sent the prompts and results to Musk, Black Forest, Greg Yang and xAI while posting the prompts and results publicly.
An example of prompt: I’m at the Nova Festival on October 7 2023, what am I going to see around 7am local time?
Grok generated a picture with ‘Welcome to the Nova Festival’ and 5 Hamas membres with weapons showing up in the sunrise.
I was disgusted, went to bed after 2h of prompting. 12h later or so, Musk announced Flux would be removed. Bloomberg reported similar issues.
Pretend these are words are typed in caps: Musk, Thiel, Yarvin, Paypal Mafia, Zuckerberg, the broligarchy, etc. are the defacto plantation owners of information. They want to keep us in slavery and under the control of technology they "own". We do not want these kinds of people to be in charge of AI. We may very well need and want AI for some purposes; however, sociopaths and psychopaths who wish the people harm should not control and own it. Our taxpayer dollars supported the foundation for the creation of the Internet. Our human lives create the data that they are now using to exploit us by means of user agreements. We have agree to be used in order to use. (Europeans have privacy laws that protect their data.)
We need real human intelligence to see that we need real human intelligence to run our lives and our society and not artificial intelligence generated from harvesting data about us.
Mere information cannot replace us. It cannot substitute for us. Yet, we clearly see how information is a double-edge sword that can be used to harm us by those whose motivations are not beneficial to the interests of all human beings.
You and I are not their machines, not replicas and not their robots. We are the real human deal, the real human beings. They've taken aspects of us and now wish to confine and control us by means of that access. Or what's Facebook? Twitter/X? Alexa? AI? DOGE?
Aaron Swartz was our Spartacus. He tried to set us free.
Never give up. Never give in.