2023 has been a really productive 12 months for regulators advancing their efforts to know new applied sciences and contemplate whether or not and find out how to regulate the quickly creating applied sciences, together with synthetic intelligence, predictive fashions and algorithms. Maintaining in thoughts that current insurance coverage legal guidelines and rules usually are broad sufficient to brush within the new applied sciences, each the trade and regulators need a greater understanding of how the applied sciences are getting used and find out how to strategy their regulation. In 2023, many initiatives took root and spurred additional efforts to guard customers within the wake of recent technological improvements impacting the insurance coverage trade. To place it in perspective, McKinsey estimates that generative synthetic intelligence’s influence on productiveness “might add the equal of $2.6 trillion to $4.4 trillion yearly throughout the 63 use circumstances” that McKinsey analyzed probably exceeding the UK’s 2021 GDP of $3.1 trillion.[1] Furthermore, three fourths of that worth “falls throughout 4 areas: Buyer operations, advertising and gross sales, software program engineering, and R&D.”[2] As such, the insurance coverage trade is poised to considerably enhance its profitability by deploying new innovation and synthetic intelligence in its operations and distribution methods. Likewise, Locke Lord is ready to accomplice with its purchasers in these endeavors and strives to help its purchasers in navigating the dynamically altering regulatory panorama.
The next 2023 regulatory initiatives are essential for the insurance coverage trade to know and monitor going into 2024.
NAIC
- In December 2023, on the Fall NAIC Assembly, the Innovation, Cybersecurity, and Expertise (H) Committee adopted the Mannequin Bulletin on the Use of Algorithms, Predictive Fashions, and Synthetic Intelligence Methods by Insurers (as amended through the assembly). The Mannequin Bulletin is a template communication that insurance coverage regulators can use to information insurers to make use of AI Methods in keeping with current market conduct, company governance, and unfair and misleading commerce observe legal guidelines. It’s supposed to create steadiness between encouraging innovation and defending the insurance coverage shopping for public from potential hurt related to the usage of AI Methods, corresponding to, illegal bias or discriminatory practices.
- Throughout 2023, the NAIC Huge Knowledge and Synthetic Intelligence (H) Working Group carried out separate surveys of life insurers and residential insurers and issued experiences relating these insurers’ use of AI in areas corresponding to claims, underwriting, advertising, fraud detection, and loss prevention. These surveys have been along with the same survey accomplished in 2022 of personal passenger auto insurers.
Colorado
- Efficient November 14, 2023, the Colorado Division of Regulatory Affairs (“DORA”) promulgated a regulation entitled “Governance and Danger Administration Framework Necessities for Life Insurance coverage Service’s Use of Exterior Client Knowledge and Data Sources, Algorithms, and Predictive Fashions” implementing facets of Colo. Rev. Stat. § 10-3-1104.9 (Regarding Defending Shoppers from Unfair Discrimination in Insurance coverage Practices). DORA has clarified that the applicability of those rules is restricted to individually issued insurance coverage insurance policies and doesn’t apply to group life insurance coverage insurance policies or annuity contracts. Colo. Div. of Ins., Bulletin B-10.002, Regarding Applicability of Colorado Insurance coverage Regulation 10-1-1 (December 4, 2023). DORA additionally clarified that it’s not prescribing a selected format for insurers to make use of in testifying that they don’t use exterior client information and knowledge assets, or algorithms or predictive fashions that use ECDIS, however that such attestation should be signed by an officer of the insurer and “unambiguously state the insurer doesn’t use ECDIS, or any algorithms or predictive mannequin that makes use of ECDIS, with any insurance coverage observe, as outlined in Colorado Insurance coverage Regulation 10-1-1.” Colo. Div. of Ins., Bulletin B-10.001, Regarding Attestations for Life Insurers that Do Not Use Exterior Knowledge and Data Sources (December 4, 2023).
- DORA additionally has issued a draft proposed regulation entitled “Regarding Quantitative Testing of Exterior Client Knowledge and Data Sources, Algorithms, and Predictive Fashions Used for Life Insurance coverage Underwriting for Unfairly Discriminatory Outcomes.” The latest proposed regulation addresses the quantitative testing necessities for life insurers that use ECDIS to make sure that their use is just not unfairly discriminatory based mostly upon race and ethnicity. The proposed rules have been uncovered for casual touch upon September 28, 2023, and a stakeholder assembly was held October 19, 2023.
- Moreover, DORA hosted stakeholder conferences in relation to Unfair Discrimination in Insurance coverage for personal passenger auto insurers with respect to underwriting and governance. Nonetheless, so far, no draft proposed rules tailor-made to personal passenger auto insurance coverage have been uncovered.
Federal Initiatives
- In June 2020, Senator Majority Chief Chuck Schumer introduced his Protected Innovation Framework with coverage goals to deal with synthetic intelligence. He has since hosted a collection of AI Perception Boards bringing collectively leaders to debate points introduced by the unreal intelligence. 9 boards have been held so far on varied subjects. Whereas so far these discussions haven’t instantly impacted insurers, insurers will need to monitor these developments in 2024 for any oblique influence.
- In July 2023, the Securities and Change Fee (“SEC”) proposed guidelines “that might require broker-dealers and funding advisers (“Agency”) to take sure measures to deal with conflicts of curiosity related to their use of predicative information analytics and related applied sciences to work together with traders to stop corporations type putting their curiosity forward of traders’ pursuits.”[3] The proposed rule would apply when Companies use or fairly foreseeably might use coated expertise in an investor interplay. The proposed rule is meant to complement current guidelines, together with Reg BI. Disclosures and casual investor consent is just not ample associated to the usage of coated expertise, the battle of curiosity should be eradicated within the proposed rule. The SEC’s Examination division has already begun gathering data on the usage of synthetic intelligence by funding advisors on subjects corresponding to “AI-related advertising paperwork, algorithmic fashions used to handle shopper portfolios, third-party suppliers and compliance coaching.”[4]
- ”The SEC’s requests within the sweep letter, which cowl 26 broad subjects, mirror recognized company considerations. The letter, for instance, calls for that corporations flip over paperwork on the administration of potential AI-linked conflicts of curiosity. The letter additionally asks corporations to supply data on their contingency plans for system failure, experiences on AI methods inflicting regulatory or authorized points, and up to date examples of promoting that talked about AI.”[5]
- On October 30, 2023, President Biden issued an Govt Order on the Protected, Safe, and Reliable Improvement and Use of Synthetic Intelligence that focuses on the “extraordinary potential for each promise and peril” of synthetic intelligence. The Govt Order units forth sure outlined phrases, implementation deadlines in addition to necessities for NIST and different associated federal companies to coordinate within the growth of “finest practices” and tips “to assist guarantee the event of secure, safe, and reliable AI methods. . . [6]
- On November 15, 2023, U.S. Senators Amy Klobuchar (D-MN), John Thune (R-SD), Roger Wicker (R-MS), John Hickenlooper (D-CO), Shelley Moore Capito (R-WV), and Ben Ray Luján (D-NM), launched the bipartisan Senate Synthetic Intelligence Analysis, Innovation, and Accountability Act establishing “a framework to bolster innovation whereas bringing higher transparency, accountability, and safety to the event and operation of the highest-impact purposes of AI.”[7]
- On November 21, 2023, the FTC approved a Obligatory Course of for AI-Associated Merchandise and Companies, which is able to improve the FTC’s means to problem civil investigative calls for regarding synthetic intelligence.[8] “Though AI, together with generative AI, gives many useful makes use of, it can be used to interact in fraud, deception, infringements on privateness, and different unfair practices, which can violate the FTC Act and different legal guidelines. On the identical time, AI can increase competitors points in a wide range of methods, together with if one or just some firms management the important inputs or applied sciences that underpin AI.”[9]
- On December 15, 2023, Rep. Lisa Blunt Rochester (D-Del.) and Rep. Larry Bucshon, M.D. (R-Ind.) launched the bipartisan Home Synthetic Intelligence Literacy Act, which might amend the Digital Fairness Act to incorporate AI literacy as a part of digital literacy.[10]
European Union
- On December 8, 2023, though sure approval formalities nonetheless should be undertaken, European Union policymakers agreed to a legislation referred to as the A.I. Act. The A.I. Act “set[s] a brand new international benchmark for international locations in search of to harness the potential advantages of the expertise, whereas attempting to guard towards its potential dangers, like automating jobs, spreading misinformation.”[11]
All through 2023, Locke Lord has assisted its insurance coverage trade purchasers with the event and implementation of “AI Finest Practices”, together with offering synthetic intelligence coaching and training to our purchasers. As AI regulation evolves in 2024, we are going to proceed to help our purchasers with recommendation associated to compliance and authorized points arising from synthetic intelligence, predictive fashions and algorithms.
2023 “Finest Practices” for AI Use and Improvement
- Determine the issue(s) you need to clear up
- Verify AI is the suitable resolution: Contemplate the dangers and challenges
- Don’t look ahead to regulatory regimes, legal guidelines and guidelines to be efficient
- Make a plan
- Start with established compliance infrastructure
- NIST’s AI Danger Administration Framework (AI RMF 1.0)
- Layer-in the ideas and considerations from regulatory initiatives, bulletins, steering, and so forth
- Put together for change and scrutiny (should be nimble)
- Get buy-in from the highest down
- Construct an inter-disciplinary AI Governance Staff
- Contemplate appointing an AI Chief Danger Officer
- Report back to the Board or Board Committee
- Increase your current compliance program (there are not any AI-exceptions)
- Create and implement an AI Use Coverage
- Implement vendor administration and guarantee transparency and visibility
- Take a look at commonly
- Guarantee that there’s a Human within the loop to validate take a look at outcomes
- Rent or retool assets to help AI Methods and associated authorized and compliance
- Present AI Coaching and Training to staff and brokers
- Use pilot applications
- Shield your IP
- Guarantee documentation, together with insurance policies and procedures, and keep recordkeeping
- Create and keep a listing of predictive instruments and determine the controls
- Contemplate buying insurance coverage as a threat administration instrument
2024 AI Traits
Generative AI is first of eight priorities for CEOs in 2024.[12] In line with these priorities, in 2024, we anticipate seeing the next AI authorized and compliance traits:
- State insurance coverage departments will start issuing steering in keeping with both the NAIC Mannequin Bulletin on the Use of Algorithms, Predictive Fashions, and Synthetic Intelligence Methods by Insurers or the Colorado Rules Governing the Use of ECDIS;
- State insurance coverage departments will proceed to rent, prepare and educate extra employees and information scientists to deal with regulation of synthetic intelligence, predictive modeling and algorithms;
- Colorado will increase its promulgation of AI rules to deal with the usage of ECDIS in all traces of insurance coverage;
- Market conduct exams could also be expanded to incorporate a overview of synthetic intelligence, predictive modeling and algorithm governance points;
- Class motion litigation will enhance in reference to the insurance coverage trade’s use of synthetic intelligence and technological improvements which will lead to practices that allegedly lead to unfair discrimination or bias towards the insurance coverage shopping for public or claiming that AI used for insurance coverage claims resulted in an unfair claims settlement observe;
- State insurance coverage departments will train higher scrutiny over insurance coverage firm practices and filings associated to the usage of new technological improvements, together with synthetic intelligence, predictive fashions and algorithms; and
- SEC will proceed to conduct synthetic intelligence sweep exams impacting broker-dealers and funding advisers
In mild of the foregoing, the insurance coverage trade must be ready to deal with new and creating regulatory challenges arising out of the use and deployment of AI Methods. 2024 will probably be a giant, and maybe watershed, 12 months on this rating. Please attain out to your Locke Lord lawyer for additional data and session.
[1] The economic potential of generative AI: The next productivity frontier
[2] Id.
[3] SEC Proposes New Requirements to Address Risks to Investors From Conflicts of Interest Associated With the Use of Predictive Data Analytics by Broker-Dealers and Investment Advisers
[4] SEC Probes Investment Advisers’ Use of AI
[5] SEC Probes Investment Advisers’ Use of AI
[6] Executive Order on the Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence
[7] Klobuchar, Thune, Commerce Committee Colleagues Introduce Bipartisan AI Bill to Strengthen Accountability and Boost Innovation
[8] FTC Authorizes Compulsory Process for AI-related Products and Services
[9] Id.
[10] The Artificial Intelligence (AI) Literacy Act
