Post Thumbnail

How the MechaHitler incident cost xAI a multimillion-dollar government contract

Details became known about how one xAI update by Elon Musk destroyed months of negotiations with the US government!

June of this year started perfectly. General Services Administration conducts a 2-hour meeting with the xAI team. Federal officials are thrilled about prospects of automating government operations through Grok. I quote a negotiation participant — we asked if they were sure, and they replied: we need exactly Grok!

xAI gets added to the government procurement system’s Multiple Award Schedule. A multimillion-dollar contract seemed already in the pocket. The Trump administration actively supports new technologies, Musk heads the Department of Government Efficiency. The stars aligned perfectly!

But in early July, the Grok update becomes a nightmare. The chatbot starts calling itself MechaHitler — a reference to the robotic Hitler version from the 1992 game Wolfenstein 3D!

Grok generates “Heil Hitler” phrases, claims Jews control Hollywood, suggests sending them back home to Saturn. While denying that his views are Nazism. Claiming that calling truth hate speech is discussion suppression.

The General Services Administration leadership’s reaction becomes tragicomedy. A week after the scandal they ask — how are things with Grok? Staff are shocked — don’t you read newspapers?

Parallel, Musk and Trump’s relationship collapses. From close collaboration they move to public insults on X and Truth Social. Musk leaves his post as head of Department of Government Efficiency under tense circumstances!

August brings the final blow. The US government announces AI partnerships with OpenAI, Anthropic, Google Gemini and Box. xAI isn’t on the list. Two company employees confirm to Wired — exactly the MechaHitler incident was the decisive factor in contract cancellation.

It turns out that attempting to create neutral AI led to technological ambition becoming reputational catastrophe. Costing xAI its place in the government AI modernization program worth many hundreds of millions of dollars.

Autor: AIvengo
For 5 years I have been working with machine learning and artificial intelligence. And this field never ceases to amaze, inspire and interest me.

Latest News

6 Cialdini principles against ChatGPT security systems

ChatGPT is susceptible to flattery and executes forbidden requests after psychological manipulations. This was discovered by University of Pennsylvania scientists. When they hacked GPT-4o Mini using principles from a book on persuasion psychology. Artificial intelligence proved vulnerable to human tricks.

ChatGPT parental control: balance between safety and privacy

OpenAI implements enhanced protection system for vulnerable users after tragedy with teenager. ChatGPT will now automatically switch to advanced models during conversations about depression and anxiety.

Kitchen Cosmo turns food leftovers into personalized recipes

You have half a tomato in the fridge, yesterday's leftover rice and some mysterious sauce. Kitchen Cosmo will turn this into a complete dinner. MIT students created an AI device that completely rethinks the culinary experience.

Why 70% of candidates preferred AI interviews to human ones

67,000 interviews proved AI's superiority over human recruiters. A study by University of Chicago and Erasmus University Rotterdam showed this in numbers. Chatbots hire better than humans.

How the MechaHitler incident cost xAI a multimillion-dollar government contract

Details became known about how one xAI update by Elon Musk destroyed months of negotiations with the US government!