Exploring the Latest in AI Impact: Insights and Legislative Leaps
Latest TL;dr looks at state responses to federal initiatives, what gov can learn from business, and managing shadow AI
When I started TL;dr in January, I made a mental note to try to do a monthly post on Artificial Intelligence (AI). The first Wednesday of the month seemed as good a time as any1, so on Feb. 7 we did a perusal of articles on the following AI subjects: trends, lessons learned and governance.
Today, we offer the March installment. Did anything interesting happen in AI over the past week? Boy howdy, did it ever. But we’re saving a deep dive on the Google Gemini cluster-flap for Friday’s GGF. Be forewarned: Opinions will be delivered.
We’ve got much more straightforward AI stuff to review today, so let’s jump right in.
States clash over what responsible AI looks like
A Feb. 21 Route 50 article discussed the varied reactions and approaches of state governments to President Biden’s executive order on AI. Of course, the issue has become partisan. “Utah Attorney General Sean Reyes and 19 other Republican state AGs sent a letter to Secretary Gina Raimondo in response to the Commerce Department’s request for information on developing guidelines to enable deployment of safe, secure and trustworthy systems,” reports Chris Teale.
Here are the top three takeaways:
Concerns Over Federal Control of AI: The Republican AGs argue Biden’s order attempts to centralize control over AI, potentially using federal oversight for political ends, including censoring content under the guise of combating disinformation. The AGs ask the administration to “work with Congress and states across the political spectrum to find bipartisan solutions that can help our country harness the power of AI and use it for the good of all, rather than only for one political party or specific groups of people.” We wholeheartedly agree that thoughtful, bipartisan oversight would be wonderful. And a pleasant surprise.
Slimming the Workforce: The article highlights the potential for AI to significantly alter the workforce, citing Oklahoma’s AI task force report. This report suggests AI could reduce the state’s government workforce by automating tasks like managing residents’ inquiries with chatbots instead of human-staffed call centers, underscoring the potential for creating more efficient delivery of government services. The article notes that outcome contradicts what many public officials have said, that AI will be used to supplement, not replace, the government workforce.
Legislative and Executive Actions on AI: Various states and cities are taking proactive steps to regulate and harness AI through legislation and executive orders. For example, California State Sen. Scott Wiener introduced a bill to establish safety standards for developing AI systems as “an opportunity to apply hard lessons learned over the last decade, as we’ve seen the consequences of allowing the unchecked growth of new technology without evaluating, understanding or mitigating the risks.” Similarly, Washington, D.C., Mayor Muriel Bowser and Alabama Governor Kay Ivey have issued orders to integrate AI into government operations responsibly, emphasizing values such as safety, accountability, and privacy. To which I say, bravo.
How Companies Are Starting to Use Generative AI to Improve Their Businesses
A Feb. 18 article from The Wall Street Journal discusses how companies are leveraging generative AI to improve efficiency, creativity, and customer engagement across various functions, from help desks to software development. Methinks there are lessons here for government leaders.
Here are three takeaways:
Significant Productivity Gains Across Functions: One company’s use of AI for summarizing customer care calls led to increased efficiency and consistency in training, with AI-generated summaries being preferred by associates in 95% of cases. That’s a big number. In specialized domains such as software development, marketing, and customer service, another company reported productivity improvements ranging from 10% to 50%. Not as big but still moves the needle.
Cost-Effective Implementation and Force Multiplication: Ally Financial reported spending about half a million dollars on building their platform and $20,000 monthly, yielding significant savings in customer care operations. Moreover, generative AI is seen as a “force multiplier for human potential,” assisting employees in becoming more productive without replacing them. “From a productivity perspective, we are saving anywhere between a minute to two minutes for customer-care calls,” said Ally’s CIO. “More importantly, we have unleashed our associates to focus on our customers, so they’re not multitasking.” For gov leaders who are experiencing massive challenges with recruiting this aspect of AI can be a game-changer.
Strategic Investment and Ambitious Goals: The necessity of strategic investment in AI technology is underscored by its potential to drive substantial advancements in productivity and innovation. “Having a strategic conversation with the senior leadership to say, the opportunity cost of not getting on to this technology as early as possible is huge,” Cisco’s CIO said. “You don’t want to miss out on maybe the most important technology of the last 100 years.”
Shining A Light on ‘Shadow AI’
While many government managers are still thinking through how to best deploy AI tools, lots of employees are already using them. A Feb. 8 article in Route 50 discusses the emerging issue of “shadow AI” in government agencies, where employees use tools like ChatGPT without formal approval, posing various risks to the agency. Here are the top three takeaways:
The Risks of Shadow AI: The unauthorized use of AI can lead to significant security and privacy risks, as sensitive data might be exposed or mishandled, something we covered in a Jan. 26 GGF post on AI. Shadow AI is challenging to control due to the easy access and popularity of tools like ChatGPT.
Employee Motivations and Consequences: Employees turn to AI to increase productivity or because they feel their agency is slow in adopting new technologies. In addition to the potential data exposure mentioned above, employees might unknowingly agree to terms and conditions not aligned with state requirements. “If you’re an individual person signing up for something and you’re clicking yes [to terms and conditions], maybe those terms aren’t actually sufficient for what your state requires,” says Amy Glasscock, program director of innovation and emerging issues at the National Association of State Chief Information Officers said. “When you say yes with your employee email, you’re agreeing to those terms for your state, not just yourself.” Hello, pink slip.
Mitigation Strategies and AI Governance: Agencies are advised to develop clear policies regarding AI use to manage the issues cited above. No need to start from scratch. The article states “governments have plenty of guidance. In December, NASCIO issued a 12-step AI blueprint for agencies starting to incorporate AI into their operations.” It also references the National Institute of Standards and Technology’s AI Risk Management Framework and the Trustworthy and Responsible AI Resource Center. If you’re not ready for a full-blown policy, at least craft an AI use statement like the one we shared in this post.
In Other’s Words
We leave you today with three quotes passed along by James Clear, author of Atomic Habits.
Roman philosopher Musonius Rufus shares a lesson on long-term thinking:
If you accomplish something good with hard work, the labor passes quickly, but the good endures; if you do something shameful in pursuit of pleasure, the pleasure passes quickly, but the shame endures.
Entrepreneur and investor Matt Griswold offers a useful approach for life — and good governance:
Think in decades and act in days.
Entrepreneur and investor Naval Ravikant on what money can’t buy:
A fit body, a calm mind, a house full of love. These things cannot be bought — they must be earned.
Onward and Upward.
The first Wednesday also happens to coincide with SGR’s series of free Wednesday Webinars focused on AI featuring Micah Gaudet, deputy city manager for the City of Maricopa, Arizona. Full disclosure: I do occasional consulting for SGR.
Will,
Confronted by the sudden emergence of AI
everyone has FOMO.
It's like the iphone, and before that the birth control pill.
Nobody has a clue what the long term consequences
are going to be of embracing it.
There's no time to figure it out!
And nobody wants to be left behind.
I commend all moral and principled persons and organizations
who have the guts and the intelligence and the will
to jump on the back of this tiger
and steer it to be a benefit not a bane to humanity.
Thank you, Will, for doing your part
by shining light on key issues
and on how city governments are addressing the challenges.
Good point about the terms-and-conditions situation. If you work in local gov and want to play around with AI, do it with a personal account, not linked to your work (government) identity.