Artificial intelligence pioneer Yoshua Bengio has warned that humanity could be losing control over increasingly powerful AI systems and urged governments and technology companies to strengthen safety measures.
Bengio, one of the researchers widely credited with helping advance modern artificial intelligence, said the rapid development of increasingly capable systems had created risks that could become difficult to manage without stronger safeguards.
In an interview with AFP, Bengio said AI development was moving quickly while researchers were still trying to understand how advanced systems behave and how they might respond to human instructions.
The Canadian computer scientist has become increasingly vocal about AI safety in recent years. He has argued that researchers should place greater emphasis on developing systems that remain controllable and aligned with human goals.
One concern is that future AI systems could become capable of planning and carrying out complex tasks with limited human supervision. Bengio said researchers needed to ensure that such systems could be reliably monitored and stopped when necessary.
He has also warned about the possibility of AI being used for harmful purposes, including cyberattacks, disinformation and the development of dangerous technologies.
The rapid expansion of AI has created a difficult balance for governments. Policymakers want to encourage innovation and economic growth while also preventing technologies from being deployed before adequate safety standards are in place.
Bengio has called for international cooperation on AI safety, arguing that countries should not allow competition between technology powers to prevent them from establishing common safeguards.
His concerns come as major technology companies continue investing heavily in increasingly advanced AI models. Companies are competing to develop systems capable of reasoning, using tools and performing increasingly complex tasks autonomously.
The debate over regulation has also intensified, with governments in Europe, North America and Asia introducing or considering new rules covering areas such as transparency, safety testing and accountability.
Bengio has supported efforts to establish independent research and testing institutions that can assess powerful AI systems before they are widely deployed.
He has also stressed that AI development should not be left solely to technology companies, saying governments, researchers and wider society should have a role in determining how the technology is developed and used.
The warning adds to growing calls from AI researchers for greater attention to long-term safety alongside the industry’s rapid technological progress.
As AI systems become more capable, the challenge for policymakers and researchers will be to ensure that innovation continues while mechanisms for human oversight, testing and control develop at the same pace.
Leave a Reply