A Mid-Year Analysis of AI

This article was originally published in Law360 on July 11, 2023 and is republished here with permission.

The past six months of 2023 have been a notable period for advancements in artificial intelligence (AI) and generative AI in particular.

This article includes a review of the implications that AI has for the legal industry, as well as analysis on how clients’ use of AI technology can affect the counsel provided by both in-house and private practice lawyers.

Since OpenAI’s ChatGPT’s release in November of 2022, generative AI specifically has been part of legal discussions surrounding the future of content.¹ Due to the revolution of open-source software, the AI space has become more accessible for the entire technology industry, as opposed to just large, established players.² As a result the AI field has boomed.

Though generative AI made up less than 1% of venture capital funding in 2022, it is likely that due to new startups and funds it will contribute a significantly larger share this year.³ This technology is forcing businesses to consider ways in which AI can make their products more efficient, as well as how it may affect industries as a whole.⁴

Even the legal industry itself has shown an interest in generative AI. One way in which AI is increasingly being leveraged is through the use of e-discovery platforms.

Although these new tools offer legal practitioners’ the potential benefits of increased efficiency and a corresponding reduction in costs, those utilizing them should maintain caution and avoid overreliance. Despite sparse regulation, users should be mindful of possible risks, such as the interaction between AI tools and privileged client information.

It is important to keep in mind that potential legal liability can arise from failing to properly review AI systems and filter privileged documents.⁵ We should expect continued investment into AI-based legal tools heading into the second half of 2023.

Beyond the legal industry, the rapid advancement of AI implementation is resulting in significant changes for companies across sectors. One industry in particular that is expected to see the heaviest investment in new AI tools is health care. More specifically, AI is beginning to revolutionize medical scanning and diagnosis, surgical assistance, and drug discovery.⁶

The rapid developments both within and outside the legal industry have led to legislative responses as AI, especially generative AI, comes with an abundance of concerns.

2022 introduced proposed legislation to an industry that was essentially lawless.

For starters, President Biden released a Blueprint for an AI Bill of Rights in October 2022 in an attempt to protect consumers.⁷ Since then, there has been a movement for AI models of certain magnitudes to be required to register, as well as obtain a license. This would allow for new models to be tested before they are released to the public.⁸

States such as California, Illinois, New York, Montana, Colorado, Virginia, Connecticut, and Maryland have already enacted legislation regarding AI.⁹

For example, in March Texas proposed HB4695, which prohibits the use of AI to provide counseling, therapy, or other mental health services unless certain criteria are met. The technology must also undergo testing and approval by the Texas Health and Human Services Commission.¹⁰ Considering that Washington, Oregon, Texas, Minnesota, and ten other states have recently proposed legislation on the matter, 2023 will very likely be a big year for AI regulation.

CEO of ChatGPT Sam Altman is also a proponent of government regulation of AI, as there are many potential harms to society that AI presents, such as jobs being replaced by technology to technology becoming so smart that its creator cannot control it.

This concern even led to more than 1,100 AI industry leaders and experts to sign a petition which called for a six-month pause on releasing new AI technology this past March.¹¹ The motivation behind this petition is the underlying fear that misinformation can be spread by those who abuse the technology. The proposed 6-month pause would prevent more harmful and unpredictable AI developments and hopefully allow more regulation to be put into place.¹² This topic is likely not going away anytime soon.

Generative AI has certainly made its mark in the first half of 2023. Developments such as Casetext CoCounsel and ChatGPT have dominated discussion within the industry, and the surrounding regulation and proposed legislation regarding AI is a hot topic. There is plenty in store for the second half of 2023.

What should legal practitioners be thinking about going into the second half of 2023?

The implications of the rising popularity of AI are twofold for legal practitioners.

First, those within the legal profession must be increasingly mindful of how AI tools are being incorporated into the practice of law. There will likely be decisions that must be made regarding whether and how to incorporate tools that assist with tasks such as discovery, legal research, and the creation of initial document drafts.

Recently, there has been discussion surrounding ChatGPT and the Model Rules of Professional Conduct, as the tool raises ethical concerns.¹³ These concerns include competency issues, security issues, as well as general fear that using a tool such as ChatGPT could lead to contention down the line.¹⁴

One cautionary tale of ChatGPT creating competency and ethical issues for legal practitioners is the recent case involving attorney Steven Schwartz of Levidow Levidow & Oberman PC. The New York personal injury firm and two of its lawyers are facing potential sanctions because of its utilization of ChatGPT,¹⁵ with the aforementioned parties submitting a brief in federal court citing six cases that were found to be nonexistent.¹⁶

Second, legal practitioners should also be aware of AI advancements within other industries, such as the health care sector, and how to advise clients on the possible legal implications of these developments. Further complicating matters, guidance in the form of regulation and legislation normally lags years behind the actual implementation of new AI tools.¹⁷ This tendency has the effect of increasing uncertainty for clients as well as the importance of receiving quality guidance from their legal advisors.

Consequently, it is important that legal practitioners maintain an awareness of their clients’ use of new AI tools, as well as keep their clients informed on possible looming regulation.

In particular, practitioners should keep an eye out for advancements at the state level due to the fact that state regulators generally move faster than their federal counterparts.

As detailed above, states that have recently introduced or passed laws focused on protecting consumers from AI related harms include California, Illinois, Texas and Colorado.¹⁸ Though the content of these states’ legislation varies, the legislation proposed is rooted in the understanding that AI technology is developing rapidly, and in order to prevent unchartered waters, regulation is the best way to move forward.¹⁹

Another consideration for practitioners is that Biden’s Blueprint for an AI Bill of Rights could be used as a framework for future regulation. Even though AI adds a significant layer of complexity, the standard Model Rules of Professional Conduct and existing legislation still apply. Staying informed will allow practitioners to provide guidance and assistance to clients looking to acquire AI dominant companies.

Specifically, legal practitioners should expect continued investment in AI tools within the health care industry, as these tools are helping providers provide real-time data and recommendations while increasing speed and accuracy of diagnosis and treatment.²⁰

However, it is important to keep in mind potential drawbacks of new AI technologies within this sector. Implicit bias within the health care industry is a problem that can lead to inequitable outcomes for people across different genders and ethnicities.²¹

Consequently, because AI technologies are developed through the information and diagnoses the provider gives, AI tools often reflect and perpetuate the same biases. For example, major algorithmic biases have been found to result from health care risk-prediction algorithms equating health care needs with costs.²² A study examining these flawed algorithms found that they led to Black patients receiving less quality care than their non-Black counterparts.²³

结语

The first half of 2023 has been important for the AI industry, with generative AI taking the lead in discussions surrounding the future of AI.

With conversations heating up around regulations of these technologies, as well as a movement to pause the continued advancement of AI, there is much for legal practitioners to think about in the remaining months of 2023.

It is also likely that practitioners will need to make decisions with respect to the implementation of new legal AI tools in their own practices while adhering to their ethical obligations.

Moreover, practitioners will need to keep themselves informed of clients’ internal AI developments to assist with advising on how these technologies should be developed to ensure that they are consistent with the existing frameworks.