US Private Company Runs Clinics for Unhoused Patients Using AI to Assist Doctors. But Is This The Right Approach?
A company called Akido Labs is running clinics in southern California, where rates of homelessness are among the highest in the nation, using artificial intelligence (AI) to assist doctors during patient visits. However, critics argue that this approach puts patients who struggle to access healthcare at risk.
The company's goal is to "pull the doctor out of the visit" by providing AI-generated diagnoses and treatment plans, which are then reviewed by a doctor. While AI can be useful in assisting medical professionals, its use in low-income clinics raises serious concerns about diagnostic accuracy and exacerbating existing health inequities.
Studies have shown that AI algorithms trained on large datasets often produce inaccurate diagnoses, particularly for patients from marginalized communities. A 2021 study found that AI algorithms under-diagnosed Black and Latinx patients more frequently than white patients, while another study published in 2024 discovered that AI misdiagnosed breast cancer screenings among Black patients at a higher rate.
Patients may not even be aware that their healthcare provider is using AI to assist with diagnoses. Medical assistants have stated that they tell their patients about the AI system listening during consultations but do not inform them of its diagnostic recommendations, which echoes an era of exploitative medical racism where Black people were experimented on without consent.
The potential impact of AI in low-income clinics goes beyond diagnostic accuracy. Advocacy groups estimate that 92 million Americans with low incomes have basic aspects of their lives decided by AI, including eligibility for Medicaid and Social Security disability insurance. Recently, federal courts have seen cases filed against large healthcare companies like UnitedHealthcare and Humana, alleging that AI systems used to decide medical coverage resulted in patients being denied care and even death.
The use of AI in healthcare disproportionately affects unhoused individuals who already face significant barriers to accessing quality healthcare. If you are financially stable, high-quality healthcare is available; however, for those struggling to get by, AI may bar them from ever receiving the care they need.
Instead of relying on AI systems that take the lead, patients and their communities should be at the forefront of healthcare decisions, ensuring that technological innovations like AI serve as tools to support human-centered care rather than replace it.
A company called Akido Labs is running clinics in southern California, where rates of homelessness are among the highest in the nation, using artificial intelligence (AI) to assist doctors during patient visits. However, critics argue that this approach puts patients who struggle to access healthcare at risk.
The company's goal is to "pull the doctor out of the visit" by providing AI-generated diagnoses and treatment plans, which are then reviewed by a doctor. While AI can be useful in assisting medical professionals, its use in low-income clinics raises serious concerns about diagnostic accuracy and exacerbating existing health inequities.
Studies have shown that AI algorithms trained on large datasets often produce inaccurate diagnoses, particularly for patients from marginalized communities. A 2021 study found that AI algorithms under-diagnosed Black and Latinx patients more frequently than white patients, while another study published in 2024 discovered that AI misdiagnosed breast cancer screenings among Black patients at a higher rate.
Patients may not even be aware that their healthcare provider is using AI to assist with diagnoses. Medical assistants have stated that they tell their patients about the AI system listening during consultations but do not inform them of its diagnostic recommendations, which echoes an era of exploitative medical racism where Black people were experimented on without consent.
The potential impact of AI in low-income clinics goes beyond diagnostic accuracy. Advocacy groups estimate that 92 million Americans with low incomes have basic aspects of their lives decided by AI, including eligibility for Medicaid and Social Security disability insurance. Recently, federal courts have seen cases filed against large healthcare companies like UnitedHealthcare and Humana, alleging that AI systems used to decide medical coverage resulted in patients being denied care and even death.
The use of AI in healthcare disproportionately affects unhoused individuals who already face significant barriers to accessing quality healthcare. If you are financially stable, high-quality healthcare is available; however, for those struggling to get by, AI may bar them from ever receiving the care they need.
Instead of relying on AI systems that take the lead, patients and their communities should be at the forefront of healthcare decisions, ensuring that technological innovations like AI serve as tools to support human-centered care rather than replace it.
. On one hand, AI can help doctors see more patients in less time, but on the other hand, it can lead to inaccurate diagnoses and exacerbate existing health inequities 
. If they're not using AI for good, it's just gonna make things worse for low-income folks who already have a tough time getting proper care 
. Patients should be at the helm, not just tech 
. What we need is more human touch, not less 
They're using AI to help doctors with diagnoses and treatment plans in clinics where people are already struggling to get care. But what if the AI is wrong? Like, really wrong? 
I've seen those studies where AI algorithms mess up diagnoses for people from marginalized communities. That's just not okay.
. And what if it affects the outcome of treatment? Like, what if someone needs meds because they've got diabetes but the AI says they don't need them? 
. I mean, think about it - these people already struggle to get their feet in the door, and then you throw AI into the mix? It's like putting a Band-Aid on a bullet wound 
. We need to be focusing on getting these folks quality care, period - not some half-baked AI system 
I'm getting really concerned about these clinics using AI in hospitals for people who are struggling to get by. Like, what if the AI makes a mistake and they end up not getting the treatment they need? 
I think we should be focusing on making sure doctors are actually paying attention to patients, not just relying on some computer program to tell them what to do. We need to make sure everyone gets quality care, no matter what their income is 
They should be told what's going on, not left in the dark while some robot is making decisions for them.
It's all about empathy and understanding, you feel? 
. And what about those who don't know they're being helped by AI? That sounds kinda weird to me 
. Plus, its not like having a human doctor is bad... I mean, what if the AI system tells the doc something wrong? Then the patient gets bad care because of it 
. Can we just have more clinics that are staffed with real doctors and nurses? That sounds like a much better idea to me 
.
like what if they don't even know they're getting a second opinion from a machine?! And those stats about AI misdiagnosing certain communities... that's just not right. I'm all for innovation and progress, but when it comes to something as important as healthcare, shouldn't we be putting people first? 
The use of AI in low-income clinics, especially for unhoused patients, is a double-edged sword 
.
. If we're going to harness the power of AI in healthcare, we need to prioritize human-centered care and make sure that patients are at the forefront of decision-making 
. No more relying on AI systems to take the lead – let's work together to create a more inclusive and equitable healthcare system that serves everyone, not just those who can afford it 
.
We need to make sure our tech is working for us, not against us... 92 million ppl rely on these systems, and it's a recipe for disaster if they're not designed with fairness in mind