education culture, innovation, resources
Turing Award Winners Andrew Barto And Richard Sutton, Highlight Dangers Of AI Development
AI industry ethics,AI safety concerns,Andrew Barto,Reinforcement learning AI,Responsible AI development,Richard Sutton,Turing Award 2025 winners
06 Mar 2025
Turing Award winners Andrew Barto and Richard Sutton warn that AI companies prioritise profit over safety, deploying AI without adequate testing. They compare this to “building a bridge and testing it by having people use it.” They stress the need for responsible AI development, echoing concerns from other leading AI experts.
Two distinguished scientists, Andrew Barto and Richard Sutton, have been awarded this year’s Turing Award for their groundbreaking contributions to artificial intelligence. The award, widely recognised as the “Nobel Prize of Computing,” honours their work in reinforcement learning—a machine learning method that enables AI systems to make optimised decisions through trial and error.
Despite this recognition, Barto and Sutton are using the occasion to highlight critical concerns about the rapid deployment of AI technologies without sufficient safety measures. Their warning echoes sentiments previously voiced by other leading AI researchers, stressing that commercial interests are being prioritised over thorough testing and responsible innovation.
Concerns over AI deployment without safeguards
Barto, a researcher at the University of Massachusetts, and Sutton, a former DeepMind research scientist, express alarm over the manner in which AI models are being introduced to the public. In an interview with The Financial Times, they compare the current industry approach to “building a bridge and testing it by having people use it.” This analogy underscores their concern that AI products are being deployed without adequate testing, potentially exposing users to unforeseen risks.
The Turing Award, often regarded as the “Nobel Prize of Computing,” comes with a $1 million prize and recognises individuals who have made lasting contributions to the field. Barto and Sutton were honoured for their foundational work in reinforcement learning, a machine learning technique that enables AI systems to learn optimal behaviours through trial and error. This method has played a crucial role in the advancement of AI, forming the basis for breakthroughs such as OpenAI’s ChatGPT and Google’s AlphaGo.
Warnings from AI experts
The issue of unsafe AI development is not new. Several renowned AI researchers, including Yoshua Bengio and Geoffrey Hinton—both previous Turing Award winners—have previously voiced similar concerns. In 2023, a collective statement from top AI researchers, engineers, and executives, including OpenAI CEO Sam Altman, cautioned that “mitigating the risk of extinction from AI should be a global priority.”
Barto reinforces these warnings, criticising AI companies for disregarding fundamental engineering principles. “Releasing software to millions of people without safeguards is not good engineering practice,” he stated.
“Engineering practice has evolved to try to mitigate the negative consequences of technology, and I don’t see that being practised by the companies that are developing.”
Reinforcement learning: A pillar of AI progress
Barto and Sutton’s work in reinforcement learning has been instrumental in shaping modern AI. The technique has played a crucial role in the development of advanced AI models, including OpenAI’s ChatGPT and Google’s AlphaGo. Google’s senior vice president, Jeff Dean, described reinforcement learning as “a lynchpin of progress in AI” and a key driver of the industry’s current trajectory.
However, despite its significance, the researchers are concerned that AI development is being driven by business motives rather than scientific responsibility. Barto criticised AI companies for prioritising commercial incentives over research, a sentiment that has been echoed by other prominent figures in the field.
AI safety: An ongoing debate among experts
As AI continues to advance, ethical concerns around its deployment extend beyond safety—tools like Portrait Pal highlight the growing debate over AI-generated content, raising questions about authorship, authenticity, and responsible AI innovation.
The call for responsible AI development is not new. Previous Turing Award winners, including Yoshua Bengio and Geoffrey Hinton—often referred to as “godfathers of AI”—have also voiced concerns over the unchecked expansion of AI technologies. In 2023, a coalition of leading AI researchers, engineers, and CEOs, including OpenAI CEO Sam Altman, issued a joint statement cautioning that “mitigating the risk of extinction from AI should be a global priority.”
Despite these warnings, AI companies continue to push forward with commercial applications at an accelerated pace. OpenAI, which has made multiple commitments to enhancing AI safety, recently faced internal challenges regarding the balance between innovation and responsible development.
The company briefly removed Altman from his leadership role, citing concerns over “over-commercialising advances before understanding the consequences.” By December, OpenAI had announced its transition into a for-profit entity, further reinforcing concerns about the prioritisation of revenue over research.
