In a demonstration of just how sensitive product information has become in the AI industry, an xAI employee resigned under threat of termination after being asked to delete a tweet that mentioned and ranked the unreleased ‘Grok 3’ AI chatbot.
The incident highlights the intense scrutiny and control AI companies maintain over their product information, even when publicly discussed by leadership.
The Incident
Engineer Benjamin De Kraker posted a tweet ranking various AI models’ coding capabilities, including the yet-to-be-released “Grok 3,” placing it below several OpenAI models but above others in the market. When asked by xAI to delete the tweet or face termination, the employee chose to resign, stating it was “a clearly labelled opinion that contains absolutely nothing controversial.”
This is the Tweet in question:
The ranking currently (my opinion), for code:
ChatGPT o1-pro
o1
o3-mini
(all kind of tied)Grok 3 (expected, tbd)
Claude 3.5 Sonnet
DeepSeek
GPT-4o
Grok 2
Gemini 2.0 Pro Series (might be higher, will probably move up)
— Benjamin De Kraker (@BenjaminDEKR) February 8, 2025
Right or Wrong? Reddit Community Divided
The situation sparked intense debate on Reddit about corporate communication policies in the AI industry. The community was split between those who saw xAI’s response as standard corporate practice and those who viewed it as overreach.
Defenders of strict communication policies pointed to common corporate practices where employees are forbidden from discussing unreleased products or making performance comparisons with competitors. They argued that such policies are necessary to protect intellectual property and maintain market position, particularly in the highly competitive AI sector.
However, the discussion also revealed significant variations in how different tech companies approach employee communications. With some members of the Reddit AI community stating how companies like Apple maintain strict control over all employee communications, even about public information. Others, like Microsoft, reportedly allow more openness, provided employees don’t share confidential information.
Several community members with tech industry experience suggested that the real issue wasn’t the mention of Grok 3’s existence, but rather the public ranking of its capabilities against competitors. Even if framed as opinion, they argued, such comparisons from an insider could impact market perceptions and valuations.
This range of perspectives reflects broader questions about employee rights and corporate control in the AI industry, where the distinction between public knowledge and trade secrets often blurs, especially when products are pre-announced but not yet released.
Musk’s Way or the Highway?
The incident is in keeping with the management at Musk-led companies. When he first acquired Twitter, Musk fired around 80% of its workforce.
In his resignation post, De Kraker characterised it as a free speech issue:
It’s very disappointing to me that a company and leaders who supposedly champion free speech and openness would try to fire a low-level employee over a clearly-labelled opinion that contains absolutely nothing controversial.
H-AI Stakes
The incident highlights the intense pressure AI companies face in managing public communications about their products.
Several Reddit users with claimed industry experience suggested a deeper context to the story, pointing to previous instances of De Kraker discussing company matters online. While these claims remain unverified, they illustrate the challenges AI companies face in controlling their public narrative.
The community discussion revealed how AI companies must balance transparency with protecting competitive advantages. In an industry where perceived technological superiority can affect investor confidence and market valuations, even casual comments about product performance can have far-reaching implications. Some users pointed out that early performance comparisons of unreleased products could particularly impact market dynamics, regardless of whether they’re framed as personal opinion.
The stakes are especially high given the fierce competition in AI development, where companies are investing billions in research and development. Many Reddit users noted that what might seem like innocent commentary to employees could be viewed as material information by competitors or markets.
Whether xAI’s actions reflect standard corporate practice or overreach remains debated, but Benjamin De Kraker’s resignation highlights a stark reality.
In today’s cut-throat AI industry, one person’s ‘free speech’ could be deemed sensitive corporate information by someone else – with potentially career-ending consequences.
