The Artificial Intelligence Podcast

The Department of Homeland Security has established the Artificial Intelligence Safety and Security Board to protect critical infrastructure from AI-related threats. The board consists of 22 initial members, including tech industry figures such as CEOs of Open AI, Microsoft, and Alphabet. President Joe Biden ordered the creation of the board as part of an executive order on AI regulation. The board's role is to develop recommendations to prevent and prepare for disruptions caused by AI in areas such as national security and public safety. It also aims to assist critical infrastructure stakeholders in responsible AI utilization. The board includes representatives from various sectors to address the multi-faceted challenges of AI safety and security.

Oracle is boosting its generative AI capabilities to compete in the cloud industry. The launch of its chatbot Chat GPT has driven demand for cloud services and data centers. Oracle is integrating generative AI features into its cloud infrastructure and applications, enhancing productivity and reducing costs. This strategic focus on AI positions Oracle to benefit from the AI revolution. While it has trailed competitors in cloud infrastructure market share, the AI boom provides an opportunity for Oracle to catch up. Its cloud offerings are appealing to governments concerned about data sovereignty, and the company is investing in expanding its cloud computing and AI infrastructure.

The second global AI Safety Summit, co-hosted by Britain and South Korea, is encountering difficulties with lower turnout and complex questions. While the first summit resulted in the "Bletchley Declaration" on AI safety, consensus on regulation has proven challenging. Thorny issues, such as copyright, data scarcity, and environmental impact, are overshadowed by existential threats. OpenAI CEO Sam Altman emphasizes the need for an energy breakthrough and seeks to raise funds for computer chip production. However, experts caution against relying solely on scientific breakthroughs and financing. With a decrease in attendance, attention has shifted from existential risks to the resources necessary for AI development. Despite these challenges, the goal of the summit is to make progress in AI safety and innovation.

A recent study published in The Lancet Digital Health reviewed the use of artificial intelligence (AI) algorithms in clinical practice through randomized controlled trials (RCTs). The study found that the use of AI in healthcare has been increasing, with some studies suggesting that AI models can perform as well as or better than clinicians. However, most of these models have not been evaluated in real-world settings, creating uncertainty about potential risks. The study identified 86 unique RCTs related to various specialties, with gastroenterology, radiology, surgery, and cardiology being the most common. The majority of trials reported significant improvements in their primary endpoints, but further research is needed to understand the true effects and limitations of AI in healthcare.

European data protection advocacy group noyb has filed a complaint against artificial intelligence company Open AI, accusing the company of violating the General Data Protection Regulation (GDPR) by failing to ensure accuracy of personal data processed by their language model, Chat GPT. Open AI has admitted that it cannot correct incorrect information generated by Chat GPT or disclose the sources of the data used to train the model. The complaint alleges that Chat GPT generated false information about individuals, violating the GDPR requirement for data accuracy. noyb is asking the Austrian Data Protection Authority to investigate and impose penalties.

The state government of Connecticut is using AI technology provided by companies like CrowdStrike and Abnormal Security to protect itself from cyber attacks. CrowdStrike's AI system utilizes machine learning to detect and respond to cyber threats in real-time, while Abnormal Security's AI system analyzes incoming emails for anomalies that may indicate sophisticated attacks. Connecticut has contracts worth $1.3 million with CrowdStrike and $372,363 with Abnormal Security. The use of AI and its potential regulation have become subjects of scrutiny as the legislative session nears its end. Governor Ned Lamont has expressed concerns about ensuring appropriate regulation without hindering innovation.

An artificial intelligence (AI) system trained with electrocardiogram (ECG) tests has been proven to reduce the risk of death among high-risk patients by 31 percent, according to a clinical trial conducted at two Taiwanese hospitals. The AI system, which analyzed the heart test results of nearly 16,000 patients, was able to identify patients at high risk of dying based on their ECG results and alert physicians for further evaluation and treatment. The study showcases the potential of AI in healthcare, suggesting that implementing similar AI systems in hospitals worldwide can significantly reduce mortality rates and improve patient outcomes.

Higher education institutions are under pressure to incorporate AI technology in classrooms and consider its impact on the workforce. Educators should prioritize the needs of students and embrace AI to promote equitable student success. AI is a system that uses machines to make predictions or decisions and has become integrated into various aspects of daily life. A survey found that AI is already impacting policies in higher education regarding teaching, learning, technology, cybersecurity, and data privacy. Institutions like Elon University and the Community College of Aurora are leading the way in guiding the implementation and use of AI in education while promoting equity and inclusion. AI has also been used in enrollment management to support student retention and completion with positive results.

Tupac Shakur's estate is considering suing Drake over his latest diss track, in which it appears the late rapper's voice was generated using artificial intelligence (AI). The estate argues that the use of Tupac's voice without consent is a violation of his right to publicity and an abuse of his legacy. They are seeking damages for the unauthorized use of Tupac's likeness and any profits from the record. The incident raises ethical questions about the use of AI in music production. More than 200 artists have signed an open letter urging AI developers and tech companies not to undermine or replace human artistic contributions.

Social media giant Meta, formerly known as Facebook, has unveiled a new approach to combatting fake news on its platforms. The company will now label a wider range of AI-generated content on Facebook, Instagram, and Threads as 'Made with AI' in an effort to protect freedom of speech while addressing misinformation. The decision to label instead of removing content comes after criticism from the Oversight Board concerning Meta's handling of a controversial AI-altered video of President Biden. The labeling system will be implemented from May 2024, allowing users time to adjust to the notifications and reducing content removal based on the existing manipulated video policy by July 2024.