AI Watch: Stanford's Annual AI Index Reveals Top Trends
This week, Stanford University's Institute for Human-Centered Artificial Intelligence (HAI) released the highly anticipated seventh edition of its annual report on AI trends ("Artificial Intelligence Index Report 2024"). The index tracks, collates, distills, and visualizes data related to AI, and it helps to make the rapidly evolving world of AI more generally accessible.
There's a lot to like in this report, including the research team's powerful mission statement:
"Our mission is to provide unbiased, rigorously vetted, broadly sourced data in order for policymakers, researchers, executives, journalists, and the general public to develop a more thorough and nuanced understanding of the complex field of AI."
This year there are more eyes on this report than ever, and I thought it would be useful to underscore three insights you won't want to miss:
- Industry is leading academia in model development: This echoes my own experience. We are seeing a new interest in investing and building AI systems by the industries that need it most. Especially in finance and healthcare, we are seeing industry drive innovation and research, while academic research is lagging. This makes sense to me, as AI is becoming more accessible and economical; companies don’t have to wait for academic research. I'm also seeing organizations invest in private research teams, rather than collaborating with academic institutions. I'm excited to report that the number of industry and academic collaborations reached an all-time high this year, and I see more of that in our future.
- AI regulation in the US is on the rise, and the US leads the world in AI Model Development: We saw a 50% jump in AI regulation in the past year. We also saw regulation appear at local, state, and federal levels. The Executive Order was released by the White House last month set some clear guidelines for government agencies, but also US companies, to help navigate the complexities arising from fast-paced of innovation. I get asked often what country is leading the race in AI. If you use the metric of models in production, the US is leading with 61 notable models being deployed.
- AI helps workers to be more productive and leads to higher quality work: This observation is a highlight for me personally. As companies begin to navigate the world of AI, they're often looking for quick wins. One way to start seeing a positive impact on your business is to educate your workforce on how to use the AI that is rapidly showing up in the tools they use every day. At the AI Leadership Institute, we did our own survey and found that 70% of workers who received training saw an immediate boost in productivity and an increase in higher value work. This is such an important step that I created an Upskilling Guide for the teams that I work with help them get a jump start on this new, but required, skillset.
Among other things, this report crystalized much of what I experienced earlier this year after participating in the "Foundation Model Scholars Program," one of the inaugural executive workshops offered by HAI. The workshop provided access to decades-old research that led to the technologies in and surrounding Generative AI, and it gave me and my team insights that allowed us to see ahead of the curve and anticipate what's coming next. It also fueled many of the successful deployments I was part of, and now we can see those patterns emerging at scale.
The report is free, and I strongly encourage you to read it.
This week I also had the chance to speak to the National Association of Broadcasters about the importance of measuring the benefits and the risks when deploying GenAI. As you can imagine, misinformation, fake news, and deepfakes were a big part of the conversations, especially in an election year. There were hundreds of new and incumbent companies sharing how they have evolved their products and services with AI. Check out the NAB keynote for a unique perspective from the world of broadcasting!
Until next time, embrace the world of learning by doing, and share what you are learning as you try out new AI features in the products you use every day! As always, connect with me on LinkedIn with any questions, feedback, or ideas for an article.
Posted by Noelle Russell on 04/16/2024