In a revealing interview on Ashlee Vance’s “Core Memory” podcast, Jerry Tworek, a former Vice President of Research at OpenAI who spent nearly seven years at the company, offered a candid assessment of the shifting dynamics in the AI industry. Tworek argues that Google’s recent AI resurgence is as much about OpenAI’s strategic missteps as it is about Google’s own achievements.
Tworek, who announced his departure from OpenAI earlier this month to “explore types of research that are hard to do at OpenAI,” stated bluntly: “Personally, what I think you should consider Google’s comeback, I think it’s OpenAI’s fumble.” His comments come at a pivotal moment in the AI race, as OpenAI CEO Sam Altman declared a “Code Red” in December 2024 amid intensifying competition from Google.
The catalyst for this competitive shift has been Google’s Gemini 3 AI model, which received widespread industry acclaim for capabilities that some observers believe have surpassed ChatGPT. While Tworek declined to specify the exact nature of OpenAI’s missteps, he emphasized that a company with OpenAI’s advantages should have maintained its lead following the groundbreaking release of ChatGPT in 2022. “If you are a company that is ahead and has all the advantages that OpenAI has you should always stay ahead,” he asserted.
Tworek credited Google with executing effectively once it began taking large language model training seriously. “Very clearly, Google started treating seriously at that moment, training large language models and, like, through OpenAI fumbling its lead, they are very, very close now in capability and in terms of models trained,” he explained. He also noted that the entire AI industry ramped up investment after OpenAI demonstrated that ChatGPT could generate substantial revenue.
The former OpenAI executive revealed that the brutal pace of the AI race has fundamentally changed OpenAI’s research priorities. The organization, which transformed from a non-profit research lab into a public benefit corporation, now places less emphasis on risky research that may not yield immediate results. “There are multiple aspects of certain things that are just hard to do in a company that has to compete in an extremely, extremely brutal and demanding race for having the best AI model in the world right now,” Tworek said.
He explained that all major AI companies face immense pressure to demonstrate user growth, fund expensive GPU infrastructure, and maintain competitive model performance simultaneously. This pressure, according to Tworek, directly impacts companies’ “appetite for risk” in pursuing innovative but uncertain research directions. OpenAI did not respond to requests for comment on Tworek’s observations.
Key Quotes
Personally, what I think you should consider Google’s comeback, I think it’s OpenAI’s fumble.
Jerry Tworek, former VP of Research at OpenAI with nearly seven years at the company, made this striking assessment on the “Core Memory” podcast, directly attributing Google’s AI resurgence to OpenAI’s strategic mistakes rather than solely to Google’s achievements.
If you are a company that is ahead and has all the advantages that OpenAI has you should always stay ahead.
Tworek emphasized that OpenAI squandered the commanding lead it established with ChatGPT’s 2022 release, suggesting the company failed to capitalize on its first-mover advantage and superior positioning in the AI race.
There are multiple aspects of certain things that are just hard to do in a company that has to compete in an extremely, extremely brutal and demanding race for having the best AI model in the world right now.
The former OpenAI executive explained how the intense competitive pressure has forced the company to reduce its focus on risky, exploratory research in favor of maintaining its competitive position, revealing the toll that commercialization takes on innovation.
That does affect somehow your appetite for risk that you are willing to take.
Tworek described how the combined pressures of demonstrating user growth, funding expensive GPU infrastructure, and competing for model superiority have fundamentally changed how all major AI companies approach research and development decisions.
Our Take
Tworek’s departure and subsequent candid comments represent more than typical post-employment criticism—they reveal systemic challenges facing AI leaders as they navigate the transition from research to commercial dominance. His observation that OpenAI has reduced risk-taking is particularly significant, as breakthrough innovations often come from exactly the kind of exploratory research that commercial pressures discourage.
The “Code Red” response from Sam Altman validates Tworek’s assessment that Google has genuinely closed the capability gap. This suggests we’re entering a new phase of AI competition where multiple companies can produce frontier models, potentially accelerating innovation through competition while also fragmenting the market.
Most tellingly, Tworek’s decision to leave OpenAI specifically to pursue research “hard to do at OpenAI” suggests that the company’s transformation into a commercial entity has created constraints that may drive away top research talent. This brain drain could compound OpenAI’s competitive challenges, creating a self-reinforcing cycle that benefits competitors like Google who can offer different research environments.
Why This Matters
This insider perspective from a senior OpenAI researcher provides rare transparency into the competitive dynamics reshaping the AI industry. Tworek’s comments signal a potential inflection point where OpenAI’s first-mover advantage from ChatGPT may be eroding due to strategic missteps and the pressures of commercialization.
The revelation that OpenAI has reduced its appetite for risky research highlights a fundamental tension in the AI industry: balancing innovation with commercial viability. As AI companies transition from research labs to revenue-generating enterprises, they may sacrifice the exploratory work that initially gave them competitive advantages.
For businesses and investors, this suggests the AI landscape remains highly fluid, with leadership positions vulnerable to rapid shifts. Google’s resurgence with Gemini 3 demonstrates that established tech giants with substantial resources can quickly close gaps when they prioritize AI development. The “Code Red” declaration from Sam Altman underscores how seriously OpenAI views this competitive threat.
More broadly, this story illustrates how the commercialization pressures and infrastructure costs (particularly for GPUs) are fundamentally reshaping AI research priorities across the industry, potentially affecting the pace and direction of AI innovation.
Related Stories
- OpenAI Lost Nearly Half of Its AI Safety Team, Ex-Researcher Says
- OpenAI’s Competition Sparks Investor Anxiety Over Talent Retention at Microsoft, Meta, and Google
- Google Founders’ $511B Fortune Soars on AI Breakthrough Success
- Big Tech’s 2025 AI Plans: Meta, Apple, Tesla, Google Unveil Roadmap
- Elon Musk’s $97B OpenAI Bid: Strategic Move or Elaborate Troll?
Source: https://www.businessinsider.com/openai-google-ai-race-fumble-gemini-2026-1