Sally Ragab, CEO @ Neunetix.
In 1969, a psychologist named Philip Zimbardo conducted a now-famous experiment. He parked two identical cars in two neighborhoods: one in the Bronx, a high-crime area, and the other in Palo Alto, a quiet, affluent community.
The car in the Bronx was quickly vandalized, stripped within hours. The Palo Alto car sat untouched. But then Zimbardo broke one of its windows. Within days, the once-pristine car was equally wrecked.
Zimbardo's "broken window theory" suggested that environments left unchecked can lead to systemic dysfunction. I see healthcare facing its own version of this phenomenon. Prior authorizations, which I see as an outdated and fragmented process, have become the "broken window" of specialty medications—neglected, frustrating and quietly draining resources.
But unlike in the 1960s, we now have tools that can help. I believe AI offers a path to not just patch the system but to reimagine it entirely.
Where It Breaks Down
Prior authorizations for specialty medications are often managed manually—often involving faxes and phone calls. Providers report spending 12 to 15 hours a week on these tasks, which can result in delayed treatment, lost revenue and burned-out staff.
For a patient waiting on chemotherapy or a biologic for Crohn's disease, that delay isn't just inconvenient—it can be life-altering.
How AI Can Change The Game
AI isn't just a buzzword in this space; it's a practical tool that's already making an impact. Here's what I witness it enabling within the healthcare space:
• Real-time eligibility checks by analyzing patient insurance details the moment a prescription is entered.
• Predictive denial prevention, flagging incomplete or noncompliant requests before they're submitted.
• Natural language processing to extract clinical data from electronic health record notes, lab results and attachments without manual entry.
• Automated submission and tracking across payer portals, with proactive alerts on missing information.
In my direct experience, I've seen AI reduce approval times from 10 to 14 days to as little as two to three days. That's not theory—that's from real clinics we've partnered with.
Making It Work In Practice
But AI doesn't work in isolation. To be successful, providers should:
1. Start with high-friction areas (e.g., oncology, rheumatology and rare disease) where prior authorization is frequent and urgent.
2. Involve clinical staff early to identify workflow pain points.
3. Choose AI tools that integrate directly into your electronic health records and don't force a new UI.
4. Review audit logs regularly and tune your AI models to local payer patterns.
Trust And Compliance
AI in healthcare raises valid questions: Will it follow HIPAA? Can we trust it to make clinical inferences? The answer is yes, but only with oversight.
The best AI tools are transparent, explainable and built with guardrails. Vendors should offer encryption, role-based access control and Systems and Organization Controls 2 (SOC 2) compliance. And just as important: Staff need training not just on how to use the tools, but on how to challenge and verify their decisions.
The Human Layer
Building off the need for your staff to challenge AI decisions, AI isn't here to replace people; it's here to take the weight off their shoulders. It's the assistant that never sleeps, never gets overwhelmed and doesn't forget payer rules.
In the context of healthcare, I see it as a force multiplier enhancing the role of pharmacists, providers and care coordinators.
When we introduced AI at a multisite provider group, the most surprising feedback wasn't just about time saved; it was how morale improved. Nurses and coordinators said they finally felt like they could focus on patients again.
Looking Ahead
The future of prior authorizations won't be about eliminating them; it will be about making them invisible. The AI systems will know what's needed before we do. They'll draft documentation, catch gaps and smooth the back-and-forth.
But it only happens if we adopt early and thoughtfully. Prior authorizations have been a source of friction for too long. AI gives us the power to fix the "broken windows"—not with duct tape but with real structural change. The next step is ours to take.
Forbes Business Council is the foremost growth and networking organization for business owners and leaders. Do I qualify?
