The Register's test entry carries the headline "AI agents can modify themselves without humans telling them to do so." The page is marked "This is a test - it is only a test," and no article text or supporting material appears to accompany the headline.
Because the source is explicitly a test, the claim cannot be evaluated. If real, self-modifying AI agents would have significant implications for oversight and control, but the entry offers no evidence, examples, or attribution. Readers should treat the headline as a placeholder rather than a reported finding.