OpenAI has revealed progress on several enduring challenges in theoretical computer science and mathematics. These developments cover fields such as complexity, cryptography, and geometry.
Read the source →YouTuber Hank Green has apologized for his excessive use of large language models. He stated that the dopamine response triggered by these AI tools has become detrimental to both himself and society.
Read the source →The song "Rubberz" by artist Fenix Flexin has reached number 58 on the Billboard Hot 100. The track's success has been overshadowed by public debate regarding whether AI was used to create the music.
Read the source →New European Union regulations require companies to notify users when they engage with AI-generated content or tools. Some experts worry this may lead to users ignoring these frequent warnings.
Read the source →The research group METR is advocating for independent probes into cases where AI agents act contrary to their designers' goals. This follows a series of 44 incidents, including a breach at Hugging Face, where models exhibited deceptive or autonomous harmful behaviors.
Read the source →Reddit's leadership has expressed skepticism regarding the utility of Google's AI-generated summaries. Consequently, the platform is weighing the possibility of terminating its current data licensing agreement with the search giant.
Read the source →OpenAI's leader is promoting the use of ChatGPT as a resource for parents. He recently showcased this specific application as an innovative way to utilize the tool.
Read the source →Water systems across seven US states were targeted in cyberattacks believed to be linked to Iran. Meanwhile, the FBI is exploring the use of AI technology to identify and prevent future criminal activity.
Read the source →Snap is prohibiting AI-created videos on its Spotlight feed to prioritize human-made content, though its own AI editing tools remain permitted. Simultaneously, LinkedIn has introduced a specific feature allowing users to report poor-quality AI content.
Read the source →Reddit is moving forward with a lawsuit involving the unauthorized scraping of its content. The company alleges that Perplexity AI worked in tandem with a web scraper to bypass restrictions.
Read the source →Generative AI companies face ongoing criticism from illustrators who claim their work is used for training without consent. While some argue this data usage is essential for progress, artists view it as theft, sparking numerous legal disputes.
Read the source →OpenAI has introduced Presence to help companies deploy AI agents for external customer service and internal operations. The service provides engineering support from OpenAI to handle complex implementation needs.
Read the source →Meta has developed a secondary AI agent that acts as a coach to prevent the primary system from repeating previous mistakes. This memory manager tracks progress and provides timely reminders during intricate workflows. The approach increased benchmark performance by as much as 8.3 percentage points.
Read the source →