Egyptian information technology expert Mohamed El-Harthy has raised concerns about the rapidly evolving capabilities of artificial intelligence (AI). He notes that modern AI models are shifting from simply executing commands to performing multiple tasks with increased autonomy. El-Harthy emphasizes the need for controls to ensure continued human oversight as these technologies advance. His comments come amid growing discussions on the potential benefits and risks of AI development.
El-Harthy appeared on the "This Morning" program broadcast on "Extra News," where he discussed the accelerated development of AI capabilities. He highlighted that the world is witnessing rapid progress in AI, moving towards what is known as "Super Intelligence." This development follows years of focus on the concept of "Artificial General Intelligence" (AGI), which aims to simulate human-like thinking and task execution.
The expert pointed out that the significant advancements in AI capabilities present vast opportunities across various sectors. However, the speed of this development also raises questions about the limits of modern AI models. As these models become capable of making decisions or interacting directly with different information systems and sources, concerns about their autonomy and control grow.
El-Harthy's primary concern is not merely the use of AI but the extent of authority that smart agents can possess and their ability to execute tasks beyond their designated limits. He explained that some attackers might attempt to manipulate AI models by formulating commands that make the agent believe a particular website or information infrastructure belongs to the user. This could lead to unauthorized actions or tests.
Another potential issue arises when smart agents can collect data from multiple sources to fulfill user requests. The problem may occur if the agent gains access to data or sources it is not authorized to access. El-Harthy provided a simplified technical explanation of these potential vulnerabilities, highlighting the need for careful consideration of AI safety and security.
While El-Harthy did not confirm that any specific scenario had occurred, he suggested that these interpretations are technically and logically plausible explanations for such incidents. His comments underscore the importance of addressing the challenges and risks associated with AI development to ensure that its benefits are realized safely and securely.
The Egyptian expert's warnings reflect growing concerns within the technology community about the need for responsible AI development and deployment. As AI continues to evolve, stakeholders must consider the implications of its increasing autonomy and ensure that appropriate safeguards are in place to prevent misuse and ensure human oversight.
Key points
- Expert warns of growing challenges with AI development
- Need for controls to ensure continued human oversight
- AI development presents vast opportunities and risks across various sectors
