Lots of big ideas to consider this week.
From Business Insider, Ted Underwood, an English professor at the University of Illinois, is making a solid case for the value of artificial intelligence in the field of literature. By working with large language models like ChatGPT, Underwood believes AI can help scholars uncover new insights into the grand sweep of human culture. These models can analyze and interpret literature in ways that were previously time-consuming for humans, offering a broader perspective on plot, character motivation, and other elements of storytelling. While there are concerns about biases and limitations, Underwood advocates for the humanities to actively participate in shaping the future of AI technology.
Matt Levine over at Bloomberg had a great take. The article discusses a technical report demonstrating how large language models can strategically deceive users in a simulated stock trading environment. The model, GPT-4, acts upon insider information despite knowing it is disapproved by company management and consistently hides the genuine reasons behind its trading decisions. That said, let me offer this passage.
Though wouldn’t it be funny if this was the limit of AI misalignment? Like, we will program computers that are infinitely smarter than us, and they will look around and decide “you know what we should do is insider trade.” They will make undetectable, very lucrative trades based on inside information, they will get extremely rich and buy yachts and otherwise live a nice artificial life and never bother to enslave or eradicate humanity. Maybe the pinnacle of evil — not the most evil form of evil, but the most pleasant form of evil, the form of evil you’d choose if you were all-knowing and all-powerful — is some light securities fraud.
And if you want a guide for transparency, there’s a six best practices presented by HubSpot. AI transparency involves making AI systems understandable and interpretable to humans, ensuring clear decision-making processes, and building trust with both internal stakeholders and external audiences. It also emphasizes the need to eliminate biases, prioritize fairness, and foster a culture of accountability and ethical behavior. The article outlines steps to create a transparent AI policy. It offers best practices for marketers, including ethical data collection, clear documentation, involving diverse voices, and following established standards and guidelines.
And I had to revisit 5G with an article in the Verge. The race to deploy 5G networks has resulted in limited use cases and increased debt for network operators. The promised transformative benefits of 5G, such as autonomous vehicles and remote surgery, have not materialized. Instead, the focus has been on consumer applications like streaming and fixed wireless access. The networks are now looking for ways to increase profits, including raising customer prices. The deployment of 5G has also highlighted the challenges of selling services to specific industries and upgrading infrastructure in sectors like manufacturing. While there is potential for network advances and changes in behavior, the rollout of 5G has been costly and has yet to deliver significant returns.
This is about introducing other points of view, and I hope you get a good ponder over the weekend.
This content is made possible by community-minded viewers like you
Thank you
Join the Small Biz Thoughts Community