Anthropic has restored access to its Claude Fable 5 AI model after export restrictions were lifted, allowing users around the world to use the model again. The company also introduced updated safety protections designed to block certain high-risk requests more effectively. These changes were added before the model returned to public use. Users on supported plans can once again access Fable 5 through Anthropic’s services.
Soon after the relaunch, many users on online forums and social media began sharing their experiences with the updated model. Some claimed that Claude Fable no longer performed as well as it did before the temporary suspension. According to these users, the model appeared to refuse more prompts, provide shorter answers, or redirect some requests to another Claude model instead.
Anthropic has explained that the restored version includes stronger safety systems developed after reviewing reports about possible ways to bypass the model’s protections. The company said the new safeguards are designed to stop risky behavior while allowing normal everyday use. If a request triggers these protections, users are informed and, in some cases, the request is handled by Claude Opus 4.8 instead.
The company has not announced that it intentionally reduced Claude Fable’s overall intelligence or general capabilities. Instead, Anthropic says the changes are focused on improving security and making safety decisions more transparent. It also apologized for an earlier approach where some restrictions were applied without clearly informing users, promising that future safety actions would be more visible.
Despite these explanations, some developers and AI enthusiasts believe the updated safeguards make the model feel less powerful during certain technical tasks. Several users reported that coding, AI research, and advanced development prompts now face stricter limits than before. These observations have led to discussions about whether stronger safety measures can affect the overall user experience.
At the same time, many users continue to report that Claude Fable performs well for general writing, programming, research, and productivity tasks. Anthropic maintains that the model still offers the same core capabilities while preventing misuse in sensitive areas such as cybersecurity and other high-risk subjects. The company says the latest protections were created after extensive testing with government and research partners.
The debate highlights a growing challenge for AI companies as they try to balance powerful capabilities with responsible safety measures. Developers often want fewer restrictions to complete complex tasks, while AI providers must ensure that advanced systems cannot easily be misused. This balance has become one of the biggest topics in the AI industry as frontier models continue to improve.
For now, Claude Fable 5 remains available globally following its return, but opinions about its performance remain divided. While some users believe the stricter safeguards have made the model feel less capable in specific situations, Anthropic continues to state that the changes are aimed at improving safety rather than reducing overall performance. The discussion is expected to continue as more users test the updated model in real-world workloads.
Stay alert, and keep your security measures updated!
Source: Follow cybersecurity88 on X and LinkedIn for the latest cybersecurity news
