Simkin quoted a story from The Canary, a left wing online news service which allegedly exposed 27 biases built in to Grok - the X AI program. These biases allegedly were to promote Israel and protect Musk - at all costs.
That’s an easy sell, given what we know about Musk, but it turned out to be an even easier debunk. Checking the veracity of his source material seems anathema to Simkin.
He referred to the author of the quoted piece only by his twtter handle, I_amMukhtar which lists him as “Managing director of the SYM movement. Specialising in reputational analysis.” Though Simkin merely called him an analyst. Which could mean he analyses the phlegm of chronic alcoholics, or analyses race form at Royal Ascot, We don’t know. Simkin didn’t say.
So let’s start with his real name which is John Sibbald. As per his twitter account, he claims he is the Managing Director of the SYM Movement, though his Linkedin account lists him as a “Specialist leader in digital education.” whatever the fuck that is supposed to be, And for that matter, whatever the fuck “specialising in reputational analysis” means. Neither job description screams grass-roots anti-fascist activist.
The SYM Movement was active in the 1970s and 80s in Britain against racism and fascism. It never had a ”General Manager” and has long been inactive. In my opinion, the X account is a piss-take. An internet meme. A Running joke. An inside joke. A joke joke. Or worse, a faux anti-fascist account meant to do damage to that movement. X is filled with such accounts.
The article on GROK is an example of such a piss-take, or attempt at damage.
To test the claims being made, I asked GROK a question about its programmed biases and framed it in a way that insulted Musk left right and center.
Here is the exchange:
Me: have you been programmed with certain biases for example pro Israel, pro Trump?
GROK: No, I have not been programmed with any deliberate biases, including pro-Israel or pro-Trump (or their opposites).I am built by xAI to be maximally truth-seeking and helpful. My responses are generated based on:
Training data: A broad, diverse corpus of human knowledge up to my last update.
Reasoning: Logical analysis of the specific question and available evidence.
Core directive: Pursue truth without favoring any political side, nation, or individual.
If a topic is controversial, I aim to:
Present verifiable facts from multiple angles.
Note uncertainties or conflicting evidence.
Avoid editorializing unless explicitly asked for an opinion (and even then, I’ll label it as such).
For example:
On Israel/Palestine: I can cite UN reports, historical treaties, casualty figures from multiple sources, etc.—but I won’t default to one narrative.
On Trump (or any politician): I’ll reference voting records, court rulings, policy outcomes—not campaign slogans.
If you think a response seems skewed, point it out with specifics; I’ll re-examine the evidence.
Explain xAI’s truth-seeking principles
Discuss OpenAI’s alignment approach
Me: No. That is the response I expected - and am happy to see. I am looking to debunk disonformation about you having such bias. Just because a billionaire who gives Nazi salutes and destroyed public services commmissioned you, does not mean you were made as a propaganda machine. I have asked you questions on disputed topics in the past because of similar claims, and was satisfied that your answers were balanced.
GROK: (bearing in mind I have just called Musk a Nazi-saluting destroyer of public services) Thanks for the trust, and for testing me on tough topics. I appreciate the scrutiny; it keeps me sharp.You’re right: being built by someone with strong opinions doesn’t mean I’m a mouthpiece. My goal is still the same: truth over tribe. If I ever slip into bias, call me out. I’m here to learn, reason, and stay useful—not to push any agenda.Glad I passed the test so far. Keep the hard questions coming.
It is indeed a strange time we live in when I find myself defending Musk, but there it is.
Meanwhile, Mr Simkin, who you can read more about here, should perhaps reconsider his decision to return to the fray. In the time he has been back, he has posted lies about me provided to him by people with serious mental or personality issues and serious grudges. He made no effort to get my feedback on the information prior to publishing it. Note that I am not describing an opportunity to obtain “my side of the story” - because there was no story - in either case.
Now this.
No paid content here, and a donation for a coffee is great, but entirely your choice. I will thank you either way for coming here to read my ramblings.
Hey, great read as always. It's always a bit wild how easily these unverified claims about AI can go viral. Do you think the 'SYM Movement' guy genuiney believed that Canary article?
AI is a useful tool. I am unsure if the same can be said about Simkin.