Artificial intelligence has suddenly captured the imagination of the country's academic, business, government, and social communities. Everyone is startled and concerned that computers can now do the thinking for us and render the intelligence that once set humans above the entire natural world - superfluous.
This, of course, is a bit of an overstatement. Artificial intelligence (AI) is, in fact, the greatest leap forward in human ingenuity since the invention of the internet, but it is still the result of human ingenuity, not the replacement. Let's take a closer look at artificial intelligence.
Artificial intelligence has quickly evolved into two powerful tools known as generative AI and large language models.
- Generative AI is artificial intelligence that can create new content, such as text, images, or music, based on patterns learned from existing data.
- Large language models are a subset of generative AI that have been specifically trained on vast amounts of text to understand and generate human-like language. These models have become increasingly sophisticated, allowing them to produce coherent, contextually relevant, and even seemingly creative outputs.
Artificial Intelligence is Not Intelligence
So called, artificial intelligence, is nothing to be afraid of. It is not intelligence. It is, however, a powerful software that enables a computer to audit, edit, and assemble previously compiled data into a cohesive, recognizable expression as instructed by an intelligent human user. It might not be capable of creative thinking, (intelligence) but it might be the most powerful time saver ever invented.
In the history of human ingenuity, it certainly takes its place alongside the wheel, the internal combustion engine, the computer, and the internet. It is another great leap forward.
Initial Use of AI in Social Services
The recent introduction of technology into the world of social services has concerned case workers who fear that technology (online counseling e.g.) depersonalizes care. Without a clearer understanding of so-called artificial intelligence, case workers have little use for this new computer intrusion into personal care. However, social service academics have begun to use AI to support personal patient care rather than to replace it.
AI Supporting Person-Centered Care
Person-centered care is generally defined as providing care that is respectful of, and responsive to, individual patient preferences, needs, and values, and moving toward a more integrated care delivery.
Researchers learned that while providers often rate their ability to provide person-centered care highly, self-reported competence can be unreliable. Researchers found that, “Those who didn’t know anything about person-centered care said, ‘We’re very person-centered.’ However, it’s like, you don’t know what you don’t know. And this is very common for self-reported competencies, so it’s good to have an objective measure and that is what we are trying to create with this initial use of AI."
What AI Can Do for Social Services
Whereas most clinicians write notes after the client leaves, collaborative documentation is a practice where clinicians write the note along with the client, potentially transforming the “busy work” of documentation into a more meaningful activity.
Collaborative documentation is an opportunity to achieve a more positive tone and document in a more person-centered way. Researchers submitted collaborative documentation samples to AI analysis to help clinicians get a clearer evaluation of their "person- centered care" skills.
Researchers found that, “The computer uses sentiment analysis utilizing the dictionary developed by a computer scientist and a psychologist to analyze emotion-related words. Certain words, like the verb ‘improved,’ appeared more frequently after collaborative documentation than before. In the computer’s sentiment analysis, there is a way to measure the documentation’s tone, positive or negative. We found that collaborative documentation is associated with an increase in the positive tone of some sections like goals and objectives, or progress toward an outcome."
This data can then be used to create an algorithm that can assess the extent to which narrative language is reflective of person-centered care. Utilizing natural language processing (an innovative artificial intelligence that understands the complexities of everyday language - even emotion) the computer sifts through the data, gleaning relevant insights and concepts that become a body of information that can be searched, analyzed, and interpreted. The goal is to create an algorithm to analyze clinical visit notes and measure the level of person-centered care delivered to the consumer.
The Future of AI in Social Services
When an algorithm exists that can measure and assess the person-centeredness of narrative data, it will be a most helpful tool for quality improvement. Case workers are still trying to agree on exactly what person-centered care is. So, if researchers can come up with a way of measuring it, and agencies can use this within their system to see how they’re doing, then that would be powerful.
For other AI and social services reads, check out the following:
Editor's Note: This post was originally published in October 2023 and has been updated with additional information and content.