Browse by Date • Publication
Tuesday
Jan022024

2024 Promises Developments in AI, Net Neutrality, Data Security and the Universal Service Fund

Technology always advances more quickly than the rules and processes designed to regulate it -- and with newer technologies making massive leaps, 2024 promises to keep all of us busy.
The potential and realized legal issues associated with AI development and deployment cut across almost every sector of the enterprise. Issues that have been around since the beginning continue to exist, most notably bias (both intentional and un-), ethical use of AI, and the enterprise reliance on AI’s capabilities to (potentially) offload data-driven analysis (including very, very sophisticated number crunching) from humans towards machines. However, two factors remain constant: First, AI-driven products and services have no inherent common sense; and second, iData in AI systems is, by definition, strictly historical and based solely on what transpired in the past. While data can be updated frequently, but the resultant AI-deiven output is still based on past results. As such, while processes that seem at first to be ripe for AI implementation, aren’t necessarily a perfect—let alone accurate—fit.

Consider this example. A week ago, I had dinner with two friends who are both physicians, one a surgeon, and one a physician-engineer who works primarily for a large university-based health system. He has developed processes for management of scarce operating room resources in the most efficient ways including, but not limited to, scheduling. The professional experience my dinner companions brought to the table led to this conclusion: as good as AI may be in looking at historical data regarding how long specific procedures will take, it cannot relied upon to create ironclad surgical scheduling because past data is not a failsafe indicator of future scenarios. Nothing (AI or human) can predict precisely--or accurately--in all circumstances what may happen when the surgeon opens up the patient and finds issues that were unanticipated. At “game time,” when the actual situation differs from what was anticipated, an AI-generated schedule will need to be revised, if not totally thrown out the window. This will have a cascading effect on subsequent surgeries that had been scheduled, to say nothing of how patient time is “scheduled” in the recovery room, or how the hospital must staff up for pre-op, operational and post-op work. This is just one example of why AI tools, with their promise of crunching large quantities of interrelated data sets are not always equipped to generate best practices, and they must be used with oversight and caution, not unlike decisions made by humans alone.

This is only one example, but its application is much more broad. If there was any question about the focus on potential issues with AI deployment, many of the largest law firms nationwide are creating AI departments to address precisely such issues. If you’ll pardon the pun, there’s blood in the water and the big firms are sensing it. Caveat emptor!

PRIVACY AND DATA SECURITY ARE ATTRACTING GOVERNMENT ATTENTION
Privacy and data security are also very critical topics for the new year. The bad guys get more creative (and effective) in stealing data every day, and the good guys need to devote additional resources to fighting them off. The sheer volume of data—including personal identifiable information—that’s “out there” is nothing short of staggering. So too are the number of cases in the pipeline waiting to be heard either at the trial court level or on appeal. Increasing numbers of states, countries and the world at large are all ramping up laws to attempt to catch up with technology. In fact, the importance of data management and protection—particularly that which is personally identifiable—is becoming not only more commonplace and rigorous, as different layers of government take slightly different approaches to the issue. During the first week of December, the EU Parliament provisionally approved its EU AI Act, which looks to protect “high risk“ AI (a term of art) containing sensitive personal information while ensuring “fundamental rights, democracy, the rule of law and environmental sustainability while boosting innovation and making Europe a leader in the field.” What’s unique here is that the rules are specific and the violations for penalties severe (read: violators’ fines will be calculated as a portion of overall corporate gross annual earnings). This act has been designed to become the global standard for AI regulation, and with its specificity with respect to obligations and enforcement, it likely will be. The AIA also defines different AI activities and classifies them and the enforcement protocols by perceived risk. Canada has its C 27, and the U.S. is working on a far-reaching federal law, but a high level of understanding of these rules, affecting not just citizens, but everyone doing business in the EU, will absolutely be required to understand and ensure compliance.

THE RETURN OF NETWORK NEUTRALITY
So long as the Democrats hold a majority on the FCC, it is very likely that net neutrality will again return to the U.S. in 2024. This popular regulation, which was abandoned during the previous administration, is expected to return in the new year in a slightly modified form, although the FCC has yet to release the final document. However, it is fair to assume that paid traffic prioritization, traffic blocking and traffic throttling will go the way of the dodo bird in the coming year. This is good news for consumers, both individual and enterprise, and less good news for providers who will again need to treat customers more equally than they have in the past few years. Personally, I can’t wait for traffic throttling to go away. Although obviously I can only read one email at once, I hate that multiple messages come in in simultaneously.

PREPARE FOR CHANGES TO THE UNIVERSAL SERVICE FUND
Lastly, there have been numerous calls for reform for the operation and mechanism that collects and allocates funds for Universal Service which, for Q1 2024, will be 34.6% of almost every quoted communications expense. This number is on top of whatever quoted price a provider has given or claims, and is subject to sales tax, making the surcharges and fees portion of a bill, whether enterprise or consumer, often much more than anticipated. Not surprisingly, as need and costs have increased, so has the USF allocation passed on to end users to cover those costs.

It's nothing new to say that technology always advances more quickly than the rules and processes designed to regulate it. But with newer and increasingly invasive technologies making massive leaps, 2024 promises to be a year that will keep all of us busy.

With best wishes for the new year.

Originally published December 29, 2023 in No Jitter 

PrintView Printer Friendly Version

EmailEmail Article to Friend

« How to Draft a Useful Data Security Policy | Main | The EU Takes the Lead in AI Regulation – But the New Rules Will Have Global Implications »