It is little known, but nevertheless, Turing has established itself as an essential player in the race for generative artificial intelligence (AI). But why, exactly?
The secret weapon in the AI coding race
Nothing predestined the young company founded in 2018 to such a destiny. Because it specializes above all in human resources, with the mission of connecting remote developers with world-renowned companies, through AI-based optimization. Objective: fill the talent gap in the technology sector.
A meeting with OpenAI at the beginning of 2022 turned his career upside down. The AI giant, which had not yet launched ChatGPT, then called on Turing engineers to actively participate in the training of its models. They are responsible for adding code elements to their training data, a major asset to improve their performance as the data available on the web dwindles. Engineers also learn coding to models.
“ We have the best infrastructure in the world for recruiting IT engineers and data scientists. And it turns out that what these models needed most to take the next step in performance was human intelligence », explains Jonathan Siddharth, co-founder and CEO of the startup, in an interview published by The Echoes.
Noting the evolution of Sam Altman’s company, Anthropic, then Google, logically turned to Turing to accelerate the development of their AI. Because the young company has a vast pool of more than 3 million verified developers, including software engineers, AI specialists and data scientists.
A future that seems bright
The Turing model relies on remote, part-time, on-demand coders, making it a flexible solution for companies that need talent quickly. Flexibility that has proven invaluable for AI companies as they need access to massive technical resources at short notice.
Jonathan Siddharth is confident for the future, with new trends within generative AI, like agents, favoring the use of its services. “ We are moving in the direction of models capable of advanced reasoning, and code and mathematics play an important role in this “, he assures. He also believes that the new training data must come from “ human intelligence »placing Turing in a more than advantageous position.
The startup, which records a monthly turnover of $20 million, still has to come up against a major rival. Scale AI, led by young billionaire Alexandr Wang, provides training data to AI companies. It is valued at $14 billion.
- Little known, the Turing startup nevertheless stands out as a cornerstone in the development of the most powerful generative AI.
- Giants like OpenAI, Anthropic and Google use its services.
- Access to training data represents a major challenge in the race for AI.
