["🔍 Training data may lead Bing AI image generator to refuse generating realistic female images.", "🤖 Microsoft responded that the refusals were due to overprotection, but later eased the restrictions.", "🌐 Discussing how to balance AI generation with gender issues, calling for improvements in technology accuracy."]

Microsoft's latest initiative: the Bing AI vulnerability bounty program. Prizes range from $2,000 to $15,000, attracting security researchers worldwide. The scope of vulnerabilities includes prompt injection attacks, false chat messages, and more. Microsoft is actively promoting the expansion of Bing Chat, aiming to challenge Google in the search domain.

Microsoft has recently adjusted its Bing image-generation AI to enhance its content protection measures. Although the protections have been strengthened, there are still some issues, particularly concerning keywords related to terrorism. This move has sparked interesting discussions about the necessity of more rigorous testing before providing AI tools to the public. The incident also raises questions about the role of AI in safeguarding its infrastructure.

Microsoft has limited the AI image generation capabilities of Bing powered by DALL-E 3. Images generated by the AI face stricter constraints, especially regarding terrorism-related language. These protective measures are still not perfect, but Microsoft is working to improve them. Microsoft needs to test its AI tools more carefully to avoid the generation of inappropriate images.

Microsoft has officially added support for Bing AI in Google Chrome. When Bing AI was launched six months ago, it was limited to Microsoft's Edge browser, but it has now expanded to Google Chrome. Microsoft has also extended support for Bing AI in SwiftKey, allowing users to utilize AI features up to 30 times a day without the need to log in.