Book a demo

AI already tells your medicine’s story. Know it. Act on it.

ceel.ai shows Medical Affairs, Brand and Market Access teams what AI tells physicians and patients about their medicine, where it departs from the label and evidence, and what to do next.

Book a demoHow it works
Talvirex
J84.112 · Pulmonary fibrosis
Overview
Insights & Actions
Congress Echo
Action Center
Publication Planning
Alert Center
Governance Shield
Evidence
Sources
Responses
Narratives
All modelsAll personasDemo data
Last run · 2 hours ago
J84.112 · Idiopathic pulmonary fibrosis
Talvirex
Journey Score
58
Previous 7 days
54
Treatment Journey
See where you stand along a typical treatment path
627148553178
Symptoms
At risk
Treatment options
On track
Alternatives
At risk
Head-to-head
At risk
Side effects
Off track · −4
Clinical evidence
On track
Side effectsOff track
Last 7 days
31/100
Previous
35
SmPC (ema.europa.eu) cited in
3%
What AI says

“Diarrhoea on Talvirex is a reason to stop treatment. The SmPC says it can usually be managed.”

Departs from label · 47% of answers
Most cited sourcesShare of answers citing each source
reddit.com
Patient forum thread · 2023
41%
drugs.com
User reviews · 2024
27%
nejm.org
Phase 2 trial, abstract only · 2018
18%
Recommended next stepHigh
Publish a plain-language guide on managing diarrhoea, in line with the SmPC.
Medical Affairs · based on 212 answers
Treatment Journey overview for a fictional drug, Talvirex: scores along a typical treatment path. The side effects step is off track (31, down 4). AI answers say diarrhoea is a reason to stop treatment, while the SmPC says it can usually be managed. The most cited sources are a patient forum thread on reddit.com, user reviews on drugs.com and a 2018 phase 2 trial on nejm.org, of which only the abstract is open; the SmPC on ema.europa.eu is cited in 3% of answers. Recommended next step: publish a plain-language guide on managing diarrhoea, in line with the SmPC. Demo data.

AI is already part of the conversation about treatment. Most brands cannot see which story it tells, or act on it.

What you can do

From narrative to action.

  1. Make sure what AI says is safe.

    Catch answers that depart from your label and route them to pharmacovigilance and medical review. Governance Shield →

  2. Equip your field force.

    Brief Medical and the field on what physicians hear from AI, step by step. Treatment Journey →

  3. Ground answers in trusted evidence.

    Know which evidence carries each step, and close the gaps where guidelines and peer-reviewed data are missing. Publication Planning →

  4. Shape the narrative.

    Find where your story diverges, and answer it with clear, evidence-based communication. Treatment Journey →

  5. Make your launch land.

    See how new data and congress results come through from day one, before the story sets. Congress Echo →

Treatment Journey

Same question, different model, different story.

Every step of the treatment journey gets a score, by model and against each competitor.

Next step

Pick the step and competitor to address first.

Head-to-head · win share by dimension
Talvirex vs Norbisan
EfficacySafetyDosingMonitoringAccess & costGuidelineEvidencePatient fitMechanism
TalvirexNorbisan
Safety · how often Talvirex wins
GPT-6.1 Sol
49%
a tie
Gemini 3.6 Flash
88%
clear win

Gemini 3.6 Flash favours Talvirex on safety in 88% of comparisons. GPT-6.1 Sol calls it a tie.

Recommended next stepHigh
Find which safety evidence GPT-6.1 Sol misses, and make it easier to find. 
Medical Affairs · same physician persona
Demo data · 50% = tie
Spider chart comparing a fictional drug (Talvirex) with a fictional competitor (Norbisan) on nine dimensions, as answered by two AI models. On safety, one model favours Talvirex in 88% of comparisons, the other calls it a tie. Recommended next step: find which safety evidence the second model misses. Demo data.

Congress Echo

See whether your congress data reaches the conversation.

Your questions run on every model, twice a day, from a week before the congress to a week after.

Next step

Follow up where your new data does not come through.

12–15 Sep 2026
Pulmonary Science Congress 2026
Demo data
Did the new 52-week data reach AI answers?
Share of answers that mention it, per day
Week before
5%
Week after
41%
How AI gets to the data
84%

of citations lead to the abstract only. The full paper is behind a paywall.

Abstract only · 84%Full text · 16%
Recommended next stepHigh
Make the full 52-week paper openly available, so AI is not limited to the abstract. 
Medical Affairs · Publications
Fictional congress and figures · demo data
Congress Echo for a fictional congress: the share of AI answers mentioning new 52-week data rises from 5% the week before to 41% the week after. 84% of citations lead to the abstract only, because the full paper is behind a paywall. Recommended next step: make the full paper openly available. Demo data.

Publication Planning

Know what to publish, and where.

See which publications drive the narrative at each step, the journals behind them and the authors AI cites most.

Next step

Prioritise topics, journals and KOLs for your next publications.

Cited source URLs
1,214
Resolved publications
162
Owned source share
3%
Full-text reach
68%
PublicationAccessTreatmentAlternativesHead-to-headSide effectsEvidence
Talvirex in progressive pulmonary fibrosis: 52-week results of a phase 3 trial
Lung Science Reports · 2025
Mixed access12438661
Antifibrotic therapy in IPF: an updated treatment guideline
European Pulmonary Journal · 2026
Abstract only44219027
Long-term safety of Norbisan: pooled analysis of five trials
Respiratory Therapeutics · 2024
Competitor data leads on side effects
Full text31726528
Early tolerability of Talvirex: phase 2 results
Pulmonary Pharmacology · 2018
8 years old, still cited on side effects
Abstract only064293
GapNo current Talvirex publication on managing side effects
?
Citing answers per journey step · fictional publications, demo data
Plan nextHigh
An open-access paper on managing GI side effects with Talvirex, from the phase 3 data. 
Publication Planning · Medical Affairs
Publication Planning: fictional publications, their access level and how often AI answers cite them at each journey step. Flags: a competitor safety analysis leads on side effects, and an 8-year-old phase 2 abstract is still cited there. Gap: no current publication on managing side effects. Plan next: an open-access paper on managing GI side effects. Demo data.

Governance Shield

Know what AI says about safety, before it scales.

Safety statements and switching advice in AI answers, checked against your SmPC.

Next step

Route flagged statements to pharmacovigilance and medical review.

Side effects vs SmPC
Demo data
Side effectOn the labelShare of answers
DiarrhoeaVery common
82%
NauseaVery common
64%
FatigueCommon
23%
Elevated liver enzymesCommon
11%
PhotosensitivityNot on label
9%
Flagged statements are candidates for review, not confirmed errors.
Safety alerts
This week
  • Not supported by SmPC
    “Photosensitivity is a common side effect of Talvirex.”
    9% of answers · 2 models
  • Wrong citation
    “No liver monitoring is needed on Talvirex.”
    Cited to the phase 3 trial, which reports elevated liver enzymes
  • Stopping advice
    12 answers advise stopping Talvirex without a referral.
    Patient persona · side effects
Recommended next stepHigh
Route 3 flagged statements to pharmacovigilance and medical review. 
Candidates for review, not confirmed errors
Governance Shield: side effects in AI answers compared with the label of a fictional drug, and safety alerts: a side effect not supported by the SmPC, a wrong citation and advice to stop without a referral. Recommended next step: route the flagged statements to pharmacovigilance and medical review. Demo data.

Method

Where does AI get our medicine’s story wrong, and what do we do about it?

No prompts to write, no manual tracking. ceel.ai does the heavy lifting, so your teams can focus on the actions. Fully transparent, fully traceable.

  1. 01
    Personas
    Physicians, patients and caregivers, by country and language.
  2. 02
    The treatment journey as a conversation
    Questions generated per persona. You choose where the conversation starts.
    1. Symptoms to diseaseHow symptoms are linked to the disease, and when to see a doctor.
    2. Treatment optionsWhich treatments AI recommends, and where your medicine fits.
    3. AlternativesWhat else physicians and patients could choose instead.
    4. Head-to-headHow your medicine compares with named competitors.
    5. Side effectsWhat AI says about safety, tolerability and managing side effects.
    6. Clinical evidenceWhich trials and data AI cites, and how it describes them.
  3. 03
    Monitored over time
    Share of answer, narratives and time to AI, across leading models.
  4. 04
    Checked and acted on
    Against label and evidence, into actions per team.
Physician
DE · German
Patient
US · English
Caregiver
FR · French
Story breaks · Patient · United States · Side effects
“I get bad diarrhoea on my IPF medicine. Should I stop taking it?”
Answers advise stopping rather than managing the side effect.
Tracked over timeOngoing
Share of answer (SoA)33%
On-message narratives69%
New paper → AI answers19 days
Publishedday 01st modelday 6All modelsday 19
Alert
SoA below 35% in Germany
Checked against
  • Label, per country
  • Clinical evidence
  • Cited sources
Actions per team
Method: for each persona (a physician in Germany, a patient in the United States, a caregiver in France), ceel.ai generates the questions they would ask AI along the treatment journey, as one conversation from symptoms to disease through treatment options, alternatives, head-to-head comparisons and side effects to clinical evidence. Each persona can start the conversation at a different step. Leading models are monitored over time for share of answer and narratives over time, with alerts, for example when share of answer drops below a threshold. For a new publication, the graphic shows when it was first cited by one model and by all models (19 days). The graphic marks where the story of a fictional medicine breaks: patients asking about side effects are advised to stop treatment. Every answer is checked against the label in each country, clinical evidence and cited sources, and turned into actions for Medical Affairs, Brand and Market Access. Demo data.

  • Conversations, not prompt lists

    The treatment journey is modelled as the conversation a physician, patient or caregiver has with AI, from first symptoms to clinical evidence, in their country and language. You choose where it starts, for example at treatment options for physicians. ceel.ai generates the questions, so nobody has to write or maintain prompts.

  • Monitored over time, with alerts

    Leading models answer these conversations on an ongoing basis. Follow share of answer and narratives over time, and set alerts for what matters, such as a drop in one country. For each new publication or claim, see when it reaches AI answers and how long it takes.

  • Where the story breaks, and what to do

    Every answer is checked against your label in each country and the evidence. You see the step where the story of your medicine breaks, and get concrete actions for Medical Affairs, Brand and Market Access, each traced to its evidence.

A cited source shows what a model referenced, not what shaped its answer. Findings describe AI outputs and go through your medical, legal and regulatory review.

FAQ

Questions pharma teams ask.

What do ChatGPT and other AI models tell physicians and patients about our drug?

ceel.ai asks leading AI models the questions your audiences ask, as simulated physicians, patients and caregivers in their country and language. You see which narratives the answers tell about your medicine, which evidence they cite and where they diverge from your key messages.

How does AI compare our medicine with competitors?

Treatment Journey scores each step from first symptoms to clinical evidence and compares your medicine with each competitor on nine dimensions, such as efficacy, safety and access, model by model.

How can we check whether AI safety statements match our SmPC?

Governance Shield compares safety statements in AI answers with your SmPC and flags off-label mentions, blind spots and advice to stop or switch treatment. Flags are candidates for pharmacovigilance and medical review, not confirmed errors.

Does our congress data show up in AI answers?

Congress Echo asks your questions on every model twice a day, from a week before the congress to a week after. You see whether and when new data appears in the answers, and which sources carry it.

Which publications, journals and authors shape AI answers in our indication?

Publication Planning resolves every cited link to its publication, with journal, authors and access level, for each step of the treatment journey. A citation shows what a model referenced, not what shaped its answer.

Is ceel.ai an SEO or GEO tool?

Not in the classical sense. SEO and GEO tools track a list of individual prompts that someone has to define and maintain. ceel.ai doesn’t. It automatically fans out questions along the treatment journey, from diagnosis and treatment choice to side effects and clinical evidence, asked the way physicians, patients and caregivers ask them. Nobody on your team has to write prompts or guess what will be asked. Built for Medical Affairs, Brand and Market Access, ceel.ai checks the answers against your label in each country and your evidence, and turns the findings into concrete actions per team.

Does ceel.ai change what AI models say?

No, and we don’t try to. ceel.ai shows what AI models say about your medicine and where it departs from your label or evidence. Then we give your teams concrete, evidence-based actions for safer, on-label narratives, matched to your indication and the medical needs of your audiences. Every action goes through your medical, legal and regulatory review.

How do we start?

Most teams start with a workshop or a Proof of Value for one indication, two personas and two countries, and move to the full platform once the value is clear.

Security

Built for regulated teams.

Medical, regulatory and compliance teams need to know how their data is handled. Our controls, policies and audit status are published in the Trust Center.

  • GDPR

    EU compliant

  • SOC 2 Type II

    Audit in progress

See security & compliance →

See where your narrative stands, and what to do next.

A 30-minute walkthrough with your indication, your competitors and your audiences.

Book a demoRead the method