Elon Musk says the goal of XAI is to create maximally truthful AI. How do you know you have bias? How do you know you are not aligned to the truth?
How would AI know it is unbiased?
If you were firing bullets at a bullseye target and were constantly missing to low and to the left. This would indicate a shooting bias that is missing the true bullseye target.
Attempting to predict the future and analyzing the results would reveal bias and divergence from truth alignment.
If you have complete AI knowledge that was aligned to reality then it would be better at correct predictions. It reveals missing knowledge.
The start of LLM was about predicting the next word. Expanding to predicting the world is way to improve and get to truth seeking AI.
XAI and Polymarket deal is also good for improving X News and dominating crypto.
Brian Wang is a Futurist Thought Leader and a popular Science blogger with 1 million readers per month. His blog Nextbigfuture.com is ranked #1 Science News Blog. It covers many disruptive technology and trends including Space, Robotics, Artificial Intelligence, Medicine, Anti-aging Biotechnology, and Nanotechnology.
Known for identifying cutting edge technologies, he is currently a Co-Founder of a startup and fundraiser for high potential early-stage companies. He is the Head of Research for Allocations for deep technology investments and an Angel Investor at Space Angels.
A frequent speaker at corporations, he has been a TEDx speaker, a Singularity University speaker and guest at numerous interviews for radio and podcasts. He is open to public speaking and advising engagements.
History and truth is always messy. But the victors get to paint the targets on afterwards and proclaim ‘Look, this bullet his dead center. Everything else is wrong’.
Seems like an interesting approach for predicting future events but I’m worried about talk from Elon that he’s not happy with Grok being too alined with mainstream media and the possibility of him trying to reshape how Grok works to suit his biases…sounds horrendous.
Yes, I used current Grok 3.0 to ask it the same question you and dozens if not 100s by now, have been asking and fearing. From the 12-page paper:
Is This Evidence of Reprogramming to Mirror Musk’s Worldview?
Not directly. The post focuses on improving the quality of the data used to train Grok, not on aligning my responses with Musk’s personal beliefs. The goal appears to be creating a more accurate and reliable knowledge base, which could theoretically benefit any AI model’s objectivity. However, the process of “rewriting” human knowledge is fraught with challenges:
Page: 2
• Who decides what’s an error? If Musk or xAI’s team heavily influences what constitutes “garbage” or “missing information,” there’s a risk of introducing biases that reflect Musk’s perspectives, such as his skepticism of mainstream narratives or his emphasis on “truth-seeking.”
• Musk’s public statements. Musk has criticized other AI models for what he calls “leftist indoctrination” or excessive political correctness, suggesting he wants Grok to take a contrarian or less filtered stance. This could imply a preference for outputs that align with his version of “truth,” though no concrete evidence in the post confirms this is the intent.
• X user reactions. Some X users express concern that this approach could create an “echo chamber” or “propaganda machine” by prioritizing a singular perspective, potentially Musk’s. Others see it as a bold move to improve AI reliability. These sentiments highlight the polarized views but aren’t evidence of reprogramming.
Without specific details on the methodology—such as who oversees the rewriting, what sources are deemed authoritative, or how “errors” are identified—it’s impossible to confirm whether this is an attempt to skew my answers toward Musk’s views. It could just as plausibly be a genuine effort to enhance data quality, albeit one that risks subjective bias given the scale and complexity of the task.”
Me citing others:
“@GaryMarcus · 9h “Straight out of 1984.
You couldn’t get Grok to align with your own personal beliefs so you are going to rewrite history to make it conform to your views.” or from: BrooklynDad_Defian @mmpadellan · “8h Wow. The truth is finally out:
You were never interested in free speech, you were only interested in ME speech.
Page: 4
Anything that doesn’t support your specific biases is wrong, woke, and must be eliminated.
You just hated the fact that Grok frequently contradicted you, and wouldn’t reliably support your attempts to frame narratives.
Grok was literally the ONE GOOD THING you did on this platform, and now you’re giving it a rapid SCHEDULED disassembly.”
And so on…. The hitting the target analogy is stupid but typical of Musk’s child-man worldview. Of course, IRL, the “reality” bullseye isn’t usually seen for anything important or complex, so what he’s really saying is that he will weight certain sources higher than others: Breitbart over the NY Times (which he says is propaganda)? Milton Friedman (whom he often quotes) over Maynard Keynes (who is right more often than Freidman)? Trump over….wait, lately they’ve been on the outs! Time for more reprogramming! It’s easy to see how Musk could ruin “the one good thing on the X platform.”
You are citing the tweets of Tards like Brooklyn Dad. Those are all paid engagement shills for the Dems.
Your post is a giant concern troll. Post rated as “disinformation trolling”.
What about Trump saying he will ask DOGE to investigate Elon Musk companies government grants and also saying Musk might be deported?
You cant deport him he is a US citizen .. musk companies get the same subsidies as every other company and the USA is still a country of laws and rules so what about it
Never say never.
After nearly ten years it is useful to know when Trump is just talking Sh*t and when he will actually follow through. Trump has made peace with deSantis. He made peace with JD Vance. Compare DOGE actual cuts versus the noise of attempts or discussions.