A series of high-profile announcements from leading artificial intelligence firms—including claims of mathematical breakthroughs and advancements in software vulnerability detection—has drawn sharp criticism from independent experts. Critics argue that these narratives, which often frame software as incipient “artificial general intelligence,” are designed to prioritize corporate marketing over scientific accuracy.
Recent incidents, such as security breaches involving models from OpenAI, Anthropic, and Meta, were frequently characterized in media reports as instances of “rogue” AI. However, cybersecurity professionals contend that these events stem from a failure by companies to implement standard security protocols rather than autonomous machine behavior. Similarly, claims regarding mathematical achievements by models like OpenAI’s Astra have faced accusations of plagiarism and research misconduct. Mathematicians have noted that these results often lack the profound novelty initially claimed by the companies.
The focus on mathematics and computer programming as benchmarks for AI capability is viewed by some as a strategic choice. These fields allow for automated verification of outputs, which simplifies the tuning of systems without the need for extensive human annotation. A group of hundreds of mathematicians recently issued a statement warning that the industry has a strong commercial incentive to exaggerate product capabilities, urging policymakers to consult independent experts rather than relying on corporate press releases.
Critics suggest that the industry’s emphasis on the existential risks of “superintelligence” serves to shift public and regulatory focus away from immediate concerns, such as the environmental impact of data centers, data privacy, and corporate accountability. By attributing agency to the software itself, companies may effectively deflect scrutiny regarding their own development practices and the ethical implications of their work.
Experts emphasize that informed decision-making requires time and a critical examination of corporate claims. They advocate for a shift in focus toward the tangible actions of the companies building these technologies rather than speculative narratives about future digital entities.
Source: MIT Technology Review
No comments yet. Be the first to share your thoughts.