Watch Our Video Summary Capturing Top News from the Last Two Weeks
This roundup is packed: a new “pay-per-inference” web standard could reset data licensing, Apple Watch gets FDA-cleared hypertension notifications, AMA unveils major 2026 CPT updates, and Norway moves to national autonomous eye screening targeting greater than 95 percent coverage in people with diabetes. Plus: Lilly’s TuneLab, Neuralink’s brain-speech trial, a Nature feature on Delphi-2M risk prediction, NHS rollout lessons, and Amazon’s AI-matched dietitian services.
Top Stories Covered in This Video
🤖 “Pay-per-output” web standard for AI crawlers (RSL) [1] [10 Sep 2025]
Context: RSL Collective (ex-Ask.com, Yahoo RSS co-creator) proposes licensing layer atop robots.txt. Early backers include Reddit, Yahoo, Quora, Medium, The Daily Beast, Fastly.
Key point: Standard supports free/attribution/subscription, pay-per-crawl and pay-per-inference; Fastly to aid enforcement; AI firms not consulted pre-launch.
Implication: Regulatory/generics: Introduces competition that may affect pricing and formulary access. (Here, read as market structure/pricing for data licensing.)
🧪 Lilly launches TuneLab discovery platform [2] [09 Sep 2025]
https://finance.yahoo.com/news/eli-lilly-launches-platform-ai-110909916.html
Context: Reuters brief; partners include insitro (model building) and Circle Pharma (oncology).
Key point: TuneLab offers AI models trained on >$1B of Lilly data; selected biotechs contribute training data in return for access.
Implication: Partnerships/BD: Signals pipeline investment and modality expansion.
🧠 FDA to weigh AI mental health devices (DHAC) [3] [US • 11 Sep 2025]
Context: Digital Health Advisory Committee meeting set for 06 Nov 2025; public docket open; background materials to post ≥2 business days prior.
Key point: Focus on access gaps vs. risks for AI chatbots/virtual therapists; framing potential regulatory pathways.
Implication: May influence prescriber choice and payer reviews pending full data.
🏥 NHS AI imaging rollout study (UCL) [4] [UK • 10 Sep 2025]
https://www.news-medical.net/news/20250910/AI-rollout-in-NHS-hospitals-faces-major-challenges.aspx
Context: Lancet eClinicalMedicine evaluation of England’s £21m chest imaging AI program across 66 trusts.
Key point: Contracting ran 4–10 months long; by Jun 2025, ~1/3 trusts not in clinical use; issues: governance, legacy IT, staff training, engagement.
Implication: Could inform practice and payer discussions; interpretation depends on study design and confounding control.
🧾 AMA CPT 2026 adds digital health & AI codes [5] [US • 12 Sep 2025]
https://www.ama-assn.org/practice-management/cpt/288-new-cpt-codes-cover-digital-health-ai-and-more
Context: 418 total changes: 288 additions, 84 deletions, 46 revisions; Category I effective 01 Jan 2026.
Key point: New AI/augmented intelligence codes (e.g., coronary plaque, perivascular fat, burn imaging, cardiac dysfunction) and RPM over 2–15 days; 10-min monthly treatment management codes.
Implication: May influence prescriber choice and payer reviews pending full data.
⌚️ FDA clears Apple Watch hypertension notifications [6] [US • 11 Sep 2025]
Context: Clearance for Cardiovascular ML-Based Notification Software (HTNF); uses optical sensor over ~30 days; not a cuff measurement.
Key point: Feature to alert suspected hypertension; available on Series 9–11 and Ultra 2–3; planned rollout in >150 countries.
Implication: DTC/telehealth: Could streamline initiation and adherence via remote prescribing and logistics.
🧿 Neuralink brain–speech trial slated for Oct [7] [19 Sep 2025]
Context: IDE pathway; company running additional trials in US, Canada, UK, UAE.
Key point: Aim: translate imagined speech to text; leadership signals future consumer use, but current trial targets severe impairments.
Implication: May influence prescriber choice and payer reviews pending full data.
📈 Delphi-2M multi-disease risk prediction (Nature) [8] [UK • 17 Sep 2025]
https://www.nature.com/articles/d41586-025-02993-x
Context: LLM adapted to forecast 1,258 diseases using UK Biobank data (n≈400k), plus lifestyle factors.
Key point: Outperformed single-disease models in many conditions; forecasts up to 20 years ahead (dataset-specific).
Implication: Could inform practice and payer discussions; interpretation depends on study design and confounding control.
👁️ Norway to deploy autonomous AI eye screening (Eyenuk) [9] [Norway • 17 Sep 2025]
Context: South-Eastern Norway RHA choosing EyeArt®, with aim to scale nationwide; Human Bytes facilitating; DIPS Arena EMR integration.
Key point: First broad national health system adoption of autonomous AI for retinal disease detection; target >95% screening in diabetes.
Implication: Access programs: May expand screening, initiation, and follow-up at scale.
🌬️ iCARE real-world COPD/asthma program results (Adherium) [10] [US • 15 Sep 2025]
https://sg.finance.yahoo.com/news/data-icare-study-hailie-smartinhaler-040500453.html
Context: PR on preliminary outcomes from Intermountain partnership (up to 2,500 pts; 4,000 sensors).
Key point: Reported higher adherence vs. norms and reductions in admissions/readmissions (details preliminary; fuller analysis pending).
Implication: Could inform practice and payer discussions; interpretation depends on study design and confounding control.
🧑⚕️ Singapore trials “AI-free” periods to counter deskilling [11] [Singapore • 15 Sep 2025]
Context: NUHS implementing AI-off intervals; NHG exploring limits; SingHealth emphasizes rigorous evaluation.
Key point: Measures include usage limits, AI-on/off modes, and performance tracking; responds to study suggesting skill erosion risks.
Implication: Access programs: May expand screening, initiation, and follow-up at scale. (Read here as structured deployment/governance to safeguard care.)
🥗 Amazon adds AI-matched dietitian services via Fay [12] [US • 16 Sep 2025]
Context: Fay becomes first nutrition provider on Health Benefits Connector; insurance-covered RDs, AI matching.
Key point: Extends Amazon’s care navigation alongside Pharmacy and benefits tools; financial terms undisclosed.
Implication: DTC/telehealth: Could streamline initiation and adherence via remote prescribing and logistics.
Why it matters
- Payment rails for AI training/use (RSL) could reshape content economics and model accountability [1].
- Practical deployment frictions (NHS study) warn against overestimating near-term clinical lift from AI without governance/IT resourcing [4].
- Coding/reimbursement updates (CPT 2026) accelerate adoption pathways for AI and RPM services [5].
- Consumer wearables gaining FDA-cleared notifications broaden front-door screening for cardiometabolic disease [6].
- National-scale autonomous screening (Eyenuk) tests AI’s capacity to close specialty gaps and standardize access [9].
📢 Stay Ahead in AI in Healthcare and Digital Health!
✅ Make sure to like, share, and subscribe to our YouTube channel to catch all future highlights. You can also check out our weekly newsletters for more in-depth stories.
✅ Contact LucidQuest at info@lqventures.com for strategic guidance on AI in Healthcare innovations and digital health research.
FAQ
What exactly is “pay-per-inference” in RSL?
It’s a licensing term where publishers can require compensation each time AI uses their content to generate a response, in addition to pay-per-crawl for training ingestion [1].
How will Lilly’s TuneLab work with partners?
Selected biotechs get access to Lilly’s trained models and contribute their own training data; initial partners include insitro and Circle Pharma [2].
Is the FDA meeting in November about approving any mental health AI?
No. DHAC on 06 Nov 2025 will discuss benefits/risks and potential pathways; it’s advisory and non-binding [3].
Do the NHS findings mean AI doesn’t help?
Not necessarily. The study highlights procurement/integration/training hurdles and slower rollout; benefits depend on real-world implementation quality [4].
When do the new CPT AI/digital health codes take effect?
Category I codes are effective 01 Jan 2026; additions include AI analysis services and shorter-duration RPM reporting [5].
Is Apple’s hypertension feature diagnostic?
No. It’s an FDA-cleared notification to prompt cuff testing/clinical follow-up, not a definitive blood pressure measurement [6].
Entities / Keywords
RSL Collective; robots.txt licensing; pay-per-inference; Eli Lilly TuneLab; insitro; Circle Pharma; FDA DHAC; NHS England AI imaging; UCL; CPT 2026; augmented intelligence codes; RPM; Apple Watch HTNF; Neuralink IDE; Delphi-2M; UK Biobank; Eyenuk EyeArt; DIPS Arena; Intermountain Health; Adherium Hailie; NUHS/NHG/SingHealth; Amazon Health Benefits Connector; Fay (dietitians).
References
- https://arstechnica.com/tech-policy/2025/09/pay-per-output-ai-firms-blindsided-by-beefed-up-robots-txt-instructions/
- https://finance.yahoo.com/news/eli-lilly-launches-platform-ai-110909916.html
- https://www.reuters.com/business/healthcare-pharmaceuticals/fda-panel-weigh-ai-mental-health-devices-2025-09-11/
- https://www.news-medical.net/news/20250910/AI-rollout-in-NHS-hospitals-faces-major-challenges.aspx
- https://www.ama-assn.org/practice-management/cpt/288-new-cpt-codes-cover-digital-health-ai-and-more
- https://cardiovascularbusiness.com/topics/clinical/hypertension/fda-clears-apple-watch-algorithm-detecting-hypertension
- https://www.engadget.com/science/elon-musks-neuralink-plans-a-brain-speech-trial-in-october-194004627.html
- https://www.nature.com/articles/d41586-025-02993-x
- https://kfor.com/business/press-releases/globenewswire/9530142/eyenuk-secures-worlds-first-national-health-system-deployment-of-autonomous-ai-eye-screening-technology/
- https://sg.finance.yahoo.com/news/data-icare-study-hailie-smartinhaler-040500453.html
- https://www.straitstimes.com/tech/healthcare-cluster-rolls-out-ai-free-periods-for-doctors-to-prevent-de-skilling
- https://www.reuters.com/business/healthcare-pharmaceuticals/amazon-ropes-fay-offer-dietitian-services-telehealth-platform-2025-09-16/