Insights from the ABA Antitrust Law Spring Meeting
The Level Legal team journeyed to Washington, D.C., for the American Bar Association’s 73rd annual Antitrust Law Spring Meeting. Attorneys gathered to discuss pressing issues facing antitrust and competition practitioners. The discussions covered everything from high-tech topics like algorithmic pricing and AI tech stacks to high-level discussions about economic theory and the future of the Department of Justice.
Here are some of our key takeaways from the sessions.
The Non-Compete Rule: Its Fallout and Future
Panelists discussed the executive order passed under the Biden Administration that effectively banned non-compete clauses for some workers. Some of the key points included:
- Non-competes have a negative impact on worker mobility and more than 10% of low-wage workers have non-competes in place, many of whom don’t hold any intellectual property.
- Non-competes can cause businesses to forego training investments. If employees can freely “walk across the street” to a competitor, businesses have less incentive to share confidential information with workers.
- The non-compete ban may become a nonissue. A current case in the U.S. Court of Appeals for the 5th Circuit could block or vacate the rule, and the FTC itself could choose to repeal or no longer defend the rule.
What About Bob? Examining Algorithmic Pricing
Moderated by Maureen Ohlhausen, Partner at Wilson Sonsini Goodrich & Rosati and former Commissioner of the Federal Trade Commission, this session dug into a term she coined while working at the FTC: the “Bob” Test.
- The “Bob” Test uses a simple analogy to help determine whether algorithmically collected information is legal: “Everywhere the word ‘algorithm’ appears, please just insert the words ‘a guy named Bob.’ … If it isn’t ok for a guy named Bob to do it, then it probably isn’t ok for an algorithm to do it either.” (“What About Bob? Revisiting the Intersection of Antitrust Law and Algorithmic Pricing in 2024,” Artificial Intelligence and Competition Policy, Maureen Ohlhausen, Taylor Owings, and Cora Allen, Wilson Sonsini Goodrich & Rosati LLP)
- The general rule that the use of public data in algorithms is OK remains sound. It is unlikely this settled law will change and it will guide decisions regarding algorithmic pricing in antitrust matters.
- There are no bright lines for what types of data can be used in an algorithm. Just because an algorithm leverages non-public information does not mean it leads to anti-competitive behavior or is harmful to consumers.
eDiscovery: Recent Scrutiny on Document Preservation
In this ethics credit session, panelists weighed in on recent controversies in the eDiscovery space, including ephemeral messages and message deletion. They offered advice for attorneys navigating these issues, which can be fraught with ethical and legal challenges around document preservation and communication.
- The group discussed a recent high-profile case involving document preservation. A large tech company fell short of its obligations to preserve internal chats by, among other things, not monitoring whether employees preserved chats when discussing the litigation.
- Benjamin Nagin, Partner at Sidley Austin, shared three key principles for document preservation: Rely on experts (but not blindly), be persistent, and marshal evidence credibly when there is a dispute.
- Technology can identify gaps at a high level. Panelists suggested running high-level analytics to find areas where data is obviously missing (e.g., does an entire month’s worth of data have no documents or messages?).
Antitrust Issues in the AI Stack
Just about every element of the AI Stack — from data assets to chip manufacturing to cloud infrastructure — presents antitrust and competition questions. Panelists looked at the rapidly growing and changing technology through an antitrust lens in this discussion.
- Current antitrust laws are sufficient to cover AI issues. This is because AI is too early in its existence — there is too much variation for legislation not to be out of date the day it is passed (e.g., Nvidia releases new GPU configurations annually at its current pace).
- More dialogue between the tech industry and regulators is needed. There is a disconnect between enforcement agencies’ understanding of AI and what they try to enforce. However, tech companies are generally willing to bring the enforcers in and teach them about AI.
- Synthetic data is becoming increasingly important. This data has the same properties as real-world data but doesn’t include the information of real data, solving privacy issues and providing data to train AI on when sufficient real-world data is lacking.
These sessions just scratched the surface of the discussions at this year’s meeting. If you’re curious about other insights from the Level Legal team, we’d love to speak with you. Contact us today.
