Concerns about the AI industry are growing as OpenAI fires safety researchers while simultaneously revealing revenue figures $20 billion lower than investor expectations.
Imagine there are ‘watchdogs’ tasked with examining whether the AI models we use every day are safe and free from hidden risks. Now, what would you think if these watchdogs were suddenly fired by their company? And what if news emerged that the company’s own report card—its revenue performance—was far lower than expected? This is exactly what has been happening recently at OpenAI, a giant in the artificial intelligence industry.
Why Does This Matter?
This incident is not just an internal conflict at one company. It raises fundamental questions about how much we can trust AI and how transparently AI companies are being operated.
From an average user’s perspective, you might think, “As long as the AI performs well, isn’t that enough?” However, AI has already deeply integrated into our daily lives. What happens if there are no ‘safety mechanisms’ in place to monitor and correct AI when it provides false information or produces unexpected, dangerous results? This event is causing significant concern because the core personnel responsible for those safety mechanisms have now left the company.
Understanding the Context
We can compare OpenAI’s current situation to the relationship between a ‘freshman’ and a ‘strict instructor.’ The massive language models (LLMs) that OpenAI develops—AI that learns from vast amounts of data to write like a human—are highly intelligent freshmen. The ‘safety researchers’ who teach and monitor these freshmen to ensure they grow in the right direction are the strict instructors.
| Recently, OpenAI fired these instructors on the grounds that they had shared confidential information with external organizations ([OpenAI Fired 3 Safety Researchers Days After… | Let’s Data Science](https://letsdatascience.com/blog/openai-fired-three-safety-researchers-one-was-its-link-to-the-investigators)). Conversely, the fired researchers are strongly refuting the company’s claims of ‘data security violations’ (Fired OpenAI Researchers Urge Board to Maintain… - Techstrong.ai). |
In this context, the ‘external organizations’ are third-party auditing bodies. The researchers had been advocating for more transparent verification of AI safety through these bodies, but the company has effectively cut ties with them, labeling their interactions as ‘leaks.’ It is akin to a school firing teachers who monitor student misconduct under the pretext of ‘breach of confidentiality.’
Where Is the Problem?
What complicates the situation further is OpenAI’s financial state. According to recent reports, the annual revenue investors expected from OpenAI was approximately $70 billion. However, it was revealed that the actual estimate is around $50 billion—a massive gap of $20 billion (OpenAI’s revenue is reportedly $20 billion less than… - TechCrunch).
This $20 billion figure is by no means small. Given a population of 50 million, every citizen would have to contribute over $500,000 to cover that amount. This revenue discrepancy sent shockwaves through the market, even causing the stock prices of AI-related companies to falter (OpenAI’s $20 Billion Revenue Discrepancy Rattles Markets as… - Techstrong.ai). Operating losses exceeding $20 billion were already confirmed in the 2025 financials (OpenAI Leak: $38B Loss on $13B Revenue, Altman Owns Zero - Business Model Analyst).
What Happens Next?
Upon leaving, the fired researchers warned, “OpenAI must not lose its capacity for strong monitoring of AI models” (3 axed employees warn OpenAI, don’t lose monitoring… - India Today). Furthermore, they are strongly recommending that OpenAI collaborate with third-party auditing organizations to build a transparent safety ecosystem (AI researcher Mikita Balesni says he believes OpenAI fired him… - Techmeme).
Ultimately, the key lies in how OpenAI can quell market anxieties regarding its revenue and simultaneously prove that it has not abandoned its commitment to safety. We must carefully observe this process of ‘securing transparency’ that is taking place behind the scenes of dazzling technological advancements. It is important for AI to become smarter, but its true value lies in its ability to function safely by our side.
References
- AI researcher Mikita Balesni says he believesOpenAIfiredhim…
- OpenAIFiredItsSafety Staff, Then CitedItsAuditors
-
[OpenAIFired3 Safety Researchers Days After… Let’s Data Science](https://letsdatascience.com/blog/openai-fired-three-safety-researchers-one-was-its-link-to-the-investigators) - “China Stealing America’s AI Secrets” -OpenAIFired… - YouTube
- FiredOpenAIResearchers Urge Board to Maintain… - Techstrong.ai
- Investors Built $70B OpenAI Revenue Estimate Using Wrong …
- OpenAI’s Revenue Appears $20 Billion Lighter Than Reported …
- OpenAI’s Leaked Financials Show the Numbers Behind the AI …
- OpenAI Leak: $38B Loss on $13B Revenue, Altman Owns Zero
- OpenAI’s revenue is reportedly $20 billion less than …
- OpenAI’s $20 Billion Revenue Discrepancy Rattles Markets as …
- What happened to OpenAI’s $20bn? Revenue scare rattles AI trade
- 3 axed employees warnOpenAI, don’t lose monitoring… - India Today
- Missed revenue targets
- Alleged leakage of confidential information
- Lack of research output
- $5 billion
- $10 billion
- $20 billion
- Halt AI development
- Collaborate with third-party auditing organizations
- Transition to a paid service model