Microsoft has launched the “AI Anthology” blog series, which features 20 enlightening essays penned by prominent scholars and professionals from diverse fields. The purpose of this anthology is to explore how artificial intelligence can be utilized for the betterment of mankind, while also highlighting potential obstacles.
By doing so, the company aims to foster a collaborative environment that will steer AI towards a future that facilitates human flourishing, by amalgamating diverse perspectives. The inaugural post of the anthology centers around the theme of “human flourishing.”
In the post, Microsoft’s Chief Scientific Officer, Eric Horvitz, declares he has been left impressed by the capabilities of GPT-4, which according to him has shown unexpected intelligence beyond its predecessors, especially GPT-3.5. However, he raised concerns about potential disruptions and adverse consequences, such as artistic creation and attribution, malicious actors, jobs and the economy, and unknown futures.
Microsoft invited 20 experts to explore GPT-4’s capabilities and consider its potential impact on humanity. The experts were given early access to the AI tool and provided case studies in education, scientific exploration, and medicine to focus on two core questions which include how technology and its successors can contribute to human flourishing and how society can guide technology for maximal benefits.
Alec D. Gallimore, Dean of Engineering at the University of Michigan, asked the latest AI model, GPT-4, about the future of engineering education and the role of engineers in society alongside AI agents. Gallimore sees potential for AI agents to help students improve their questioning, evaluation, and checking skills. However, AI could also widen educational gaps, based on language fluency and test scores. Gallimore sees tremendous potential if educational approaches evolve to incorporate AI technology.
According to Professor of Law and Economics at the University of Toronto, Gillian Hadfield, there is a potential risk of GPT-4 reproducing online errors, generating fictitious cases and principles of law, and creating jurisdictional confusion. However, with appropriate training on factual cases and statutes in relevant jurisdictions, followed by cross-checking and citing reliable up-to-date legal sources, GPT-4 holds potential to provide legal assistance to individuals within the constraints of the law.
Ada Palmer, a distinguished professor of History at the University of Chicago, notes that AI should not be viewed as a hindrance to human flourishing. Rather, she considers it a valuable resource that has the potential to enhance students’ knowledge of civil rights and encourage them to pursue more ambitious academic projects beyond the conventional college essay.
As per Professor Robert M. Wachter, who is the Chair of the Department of Medicine at the University of California, San Francisco, the development of GPT-4 is likely to result in greater advantages than drawbacks for the healthcare industry. He harbors no doubts regarding its potential in automating complex processes such as billing and coding, documentation, patient and clinician scheduling, and supply chain management.
Meanwhile, a statement released by the Center for AI Safety features a call to action from numerous leaders of the AI industry, academics, and notable public figures to address the imminent threat posed by Artificial Intelligence.
The statement emphasizes that the risk of an AI-induced extinction event should be treated with equal gravity as other large-scale hazards like pandemics and nuclear war. The group urges the development of urgent measures to mitigate this risk and safeguard humanity’s future.