Introducing HeaRT: SSA’s New AI System for Taxpayer Support—Here’s When It Goes Live

AI has reached every level of society and is here to stay. It no longer just helps us solve newspaper puzzles, write prescriptions, or help our children with their homework.

Now, the Social Security Administration (SSA) is also going to rely on Artificial Intelligence to digitize all of its services. Don’t worry, it won’t fill anything out for you or make your life more complicated—quite the opposite! It will help the hearing and transcript systems improve process efficiency and thus make life a little easier for its program beneficiaries.

What is the SSA’s AI about?

This system they are trying to implement is called HeaRT (Hearing Recording and Transcription). It’s quite interactive, according to its acronym, and will replace the technology used until now to record and transcribe Social Security hearings, which, until now, was quite obsolete.

When will it be available?

This technology is expected to be implemented in all administrations starting March 17 to streamline and prevent errors.

What is the HeaRT system?

Simple: an Artificial Intelligence system designed to manage Social Security hearings. It will assist in the transcription of:

1. What is said in hearings without the need for physical recording and transcription equipment.
2. Hearings conducted by phone or video calls, so that the process is accurate and much faster.
3. The SSA’s current hardware, which is much more expensive and requires technical maintenance, will be eliminated.

Why?

Well, to save money! It is estimated that $5 million a year is spent on this type of equipment, and with Artificial Intelligence, that would be $5 million annually that could be invested elsewhere.

Are there benefits with this implementation?

We remember that we are increasingly moving towards a much more technological world, and that digitizing administrative processes will make everything more dynamic, accessible, and efficient.

1. The HeaRT system will allow hearing transcripts to be processed in real time, reducing wait times for applicants.
2. AI-generated transcripts minimize interpretation errors, ensuring that records are more faithful to what was actually said at the hearing.
3. Significant savings for public administration.
4. The use of AI ensures that all hearing records are accessible more quickly, which can facilitate appeals and case reviews.

But, what about the risks?

As with all technology, there are potential risks, such as the possibility that transcription systems can make mistakes, especially if there are strong accents or complex technical language, or that non-native English speakers may mispronounce some words. Professor Daniel Ho, a member of the US government’s Artificial Intelligence Advisory Committee, warned that AI can still generate errors (or “hallucinations” in technical terms), so manual review before approving official transcripts is crucial.

On the other hand, some SSA employees could see their duties reduced or even replaced, although the SSA has noted that the staff will be essential for monitoring and validating records.

Beyond the SSA.

This system will be launched on March 17, but what’s next?

1. Possible expansion of AI into other areas of Social Security, such as benefits processing and claims automation.
2. Monitoring the performance of this system, performing calculations and analyzing its efficiency and reducing errors.
3. Possible improvements in speech recognition and transcription errors.

It’s clear that AI is here to stay, and we need to embrace it as another ally that will make our lives easier, not as a giant ogre to be beaten. For now, we’ll have to see how it performs in the SSA and whether it’s truly worth using this technology in these areas or whether it’s much more comfortable to develop it in the traditional way. Will it be a step forward or a step backward?