San Francisco: A fresh debate in the artificial intelligence (AI) industry has emerged after an employee of OpenAI claimed that Google had developed a chatbot similar to ChatGPT well before its public release but chose not to launch it.

The claim was made by Thibault Sottiaux, also known as “Tibo”, who previously worked with Google’s AI division before joining OpenAI. His remarks have reignited discussions about missed opportunities and strategic caution within one of the world’s leading technology firms.

Claim of an early chatbot within Google

Responding to a post on the social media platform X, Sottiaux revealed that he was part of a team at Google that had built a ChatGPT-like system nearly a year before OpenAI launched its chatbot in November 2022.

According to him, the internal project was referred to as “LMChat” along with other codenames. He suggested that the system had capabilities comparable to what users later experienced with ChatGPT, indicating that Google may have had an early lead in conversational AI.

The statement aligns with earlier remarks by Jeff Dean, who had acknowledged in past discussions that the company possessed similar technology but did not consider it significantly better than traditional search at the time.

Concerns over impact on search business

Sottiaux further claimed that Google’s hesitation to release the chatbot stemmed from concerns about its potential impact on the company’s core search business. He stated that internal teams, including those at Google DeepMind, were restricted from launching products that could disrupt existing services.

This cautious approach reflects a broader dilemma faced by large technology companies: balancing innovation with the risk of undermining established revenue streams. Search remains a major source of income for Google, and a conversational AI interface could fundamentally change how users access information.

Google’s role in foundational AI technology

Despite the claims, it is widely recognised that Google has played a foundational role in the development of modern AI systems. The company introduced the Transformer architecture through its 2017 research paper, “Attention Is All You Need,” which underpins today’s large language models.

This architecture forms the basis of systems like ChatGPT, where “GPT” stands for Generative Pre-trained Transformer. Over the years, Google has also developed advanced models such as LaMDA, PaLM and Meena, demonstrating its leadership in AI research.

However, many of these innovations remained internal or were released cautiously, often due to concerns about accuracy, safety and ethical implications.

OpenAI’s launch changed the industry

The public release of ChatGPT by OpenAI in 2022 marked a turning point in the AI landscape. The chatbot quickly gained global attention, reaching 10 lakh users within five days and 10 crore users within two months, making it one of the fastest-growing consumer applications in history.

Following this, Google accelerated its own AI efforts, initially launching Bard and later introducing more advanced systems like Gemini. The competition between OpenAI and Google has since intensified, with both companies racing to develop more powerful and reliable AI tools.

Industry observers note that OpenAI’s willingness to release its product early, despite known limitations such as hallucinations, gave it a first-mover advantage and reshaped user expectations around AI interaction.

Safety concerns and cautious rollout

At the time, Google executives, including CEO Sundar Pichai, had emphasised the risks associated with deploying conversational AI at scale. Concerns included misinformation, hallucinations and the potential erosion of user trust.

Jeff Dean had also highlighted that while the technology met a clear user need, it was important to acknowledge its limitations. These concerns contributed to Google’s measured approach, even as competitors moved more aggressively.

A missed opportunity or strategic choice?

The revelations by Sottiaux have sparked debate over whether Google missed a significant opportunity by not launching its chatbot earlier. Some experts believe that an earlier release could have positioned Google as the leader in consumer-facing AI, potentially altering the competitive landscape.

Others argue that the company’s cautious stance helped it avoid reputational risks associated with early-stage AI deployment. Even today, issues such as inaccurate responses and hallucinations remain challenges for AI systems across the industry.

Conclusion

The claim that Google had a ChatGPT-like system before its public debut highlights the complex interplay between innovation, risk management and business strategy in the tech industry. While OpenAI capitalised on early deployment to capture global attention, Google’s cautious approach underscores the challenges of balancing technological advancement with responsibility.

As AI continues to evolve rapidly, the episode serves as a reminder that timing, execution and risk appetite can be just as critical as technological capability in determining industry leadership.