September 15, 2023

September 14, 2023

September 16, 2023

  • Healthcare

    AI can help screen for cancer—but there’s a catch

    MIT Technology Review, 09/15/23. The use of AI-assisted screening for cancer has raised concerns about overdiagnosis. Screening programs have undoubtedly saved lives by detecting cancers early. However, screening also poses risks such as infections from biopsies, complications from colonoscopies, and side effects from treatments like radiation and chemotherapy. Adewole Adamson, a dermatologist and researcher, believes that AI-assisted screening may lead to more overdiagnosis because it is challenging to differentiate lethal and nonlethal cases. Although AI may help address overdiagnosis by analyzing patients’ cancer trajectories, it is crucial to approach claims about AI cancer screening with skepticism and consider more conservative treatment options for ambiguous cases. READ THE ARTICLE

  • Art

    Funky AI-generated spiraling medieval village captivates social media

    arsTechnica, 09/15/23. A Reddit user named “Ugleh” recently posted an AI-generated image of a spiral-shaped medieval village that gained significant attention on social media. The image was created using Stable Diffusion and ControlNet techniques. Reactions to the artwork varied, with some expressing wonder and amazement, while others criticized certain aspects. Despite the attention received and offers to turn the artwork into NFTs, Ugleh has chosen to remain low key. The artwork raises questions about copyright issues and the boundaries of AI-generated art. It remains to be seen how the law will recognize the human inspiration behind such works. READ THE ARTICLE

  • Risks

    We Can Prevent AI Disaster Like We Prevented Nuclear Catastrophe

    Time, 09/15/23. The development of artificial general intelligence (AGI) poses a significant threat to humanity, with the potential to be as destructive as nuclear weapons. Just as the creation of nuclear weapons led to the establishment of international institutions like CERN and the International Atomic Energy Agency, there is a growing consensus for the creation of an international organization to deal with the risks of AGI. MAGIC (Multilateral AGI Consortium) could be the world’s only advanced and secure AI facility focused on safety-first research and development. By placing AGI development in the hands of an international organization, we can ensure a safer future for everyone. READ THE ARTICLE

  • OpenAI

    What is ChatGPT Enterprise? Business-focused Generative AI

    UC Today, 09/15/23. ChatGPT Enterprise, the upgraded version of ChatGPT, offers businesses enhanced security, privacy, and performance. With features like an administrative console, domain verification, and analytics dashboard, companies can easily manage app usage and access control. The solution addresses concerns around data safety by encrypting all input and promising not to use it for training. The performance of ChatGPT Enterprise is faster and more powerful, with no usage caps and an extended context window. OpenAI plans to introduce more features and customization options for different business roles in the future. READ THE ARTICLE

  • Journalism

    Microsoft calls deceased NBA player ‘useless’ in AI-written obituary

    Search, 09/15/23. Microsoft faced criticism after an AI-generated obituary for NBA star Brandon Hunter described him as “useless.” The offensive headline caused outrage among fans, highlighting the dangers of relying solely on AI for content generation. The subsequent article was also incoherent, further damaging Microsoft’s reputation. This incident emphasizes the importance of human supervision when using AI to create content, as factual inaccuracies and problematic errors can harm a brand’s image and negatively impact search rankings. Microsoft has removed the article but has yet to apologize officially. Other brands have also faced similar issues with AI-generated content, highlighting the need for caution and human oversight. READ THE ARTICLE

  • Jobs

    The UAW and Other Unions Must Focus More on AI and Automation in Their Negotiations

    Harvard Business Review, 09/15/23. The United Auto Workers (UAW) and other trade unions need to prioritize addressing the impact of AI and automation on jobs. While maximizing pay increases is important, it is short-sighted to overlook the changing nature of the workplace. The emergence of AI-driven technologies requires new approaches to labor relations and proactive measures to prepare workers for the future. Both unions and employers should be concerned about worker engagement and the potential disappearance of entry-level jobs. It is crucial to train workers to work alongside AI systems and develop frameworks that allow for flexibility and adaptation. The time to address these issues is now, as the transition to an AI-driven workplace will not happen overnight. Labor unions should use their leverage to ensure the future of their members before AI becomes even more pervasive. READ THE ARTICLE