The FDA can now see trial data as it emerges. Most trials still take months to start, then lose months again every time the protocol is amended. Those gaps are where development time is lost, and it is where operational AI makes the biggest difference.
Real time trials have arrived
In April 2026 the FDA announced two proof of concept real time clinical trials, run by AstraZeneca and Amgen, that report endpoints and data signals to the agency as the trial progresses. Commissioner Marty Makary described the end goal as real time, continuous trials across all phases of drug development, removing the pauses between phases.
For biotech CEOs, this raises a sharper question. If the regulator can review your data in days, where do the months in your development timeline actually go?
Manual content is the roadblock
The bottleneck sits before the first patient is screened. Tufts CSDD found that the journey from site identification to study startup completion averages 31.4 weeks, and that 61% of that cycle time is consumed by startup activities.
Manual content generation is the roadblock in every workflow, for every stakeholder. Sites wait 6 to 20+ weeks just to get the inputs they need from sponsors or CROs. They then put in a further 21+ hours of manual work per complex study to create the documents they need on site.
Much of that time goes to interpreting the protocol: reading through the Schedule of Activities, footnotes, visit windows, eligibility criteria and procedures to work out what has to be captured at each visit. Only then can sites build the visit worksheets, source documents, consent materials, lab listings and training their teams work from.
Real time reporting does nothing to shorten this, because real time reporting only begins once the trial is running.
Then the amendments arrive
Starting slow is only half the problem. Tufts CSDD research published in 2024 found that 76% of protocols now have at least one amendment, up from 57%, and the average number of amendments per protocol has risen to 3.3. Sponsors classed 77% of them as unavoidable, with regulatory agency requests a leading cause.
An amendment rarely stops a trial outright, but it stalls the work it touches. The time from identifying the need to amend to the last oversight approval now averages 260 days, and investigative sites operate on different protocol versions for an average of 215 days. Every affected document must be rewritten, reviewed, approved and redistributed, and enrolled patients may need to be reconsented. The time from final internal approval to the first patient reconsented has reached 89 days.
Real time trials will surface safety and efficacy signals sooner, and earlier signals mean earlier decisions to amend. A trial can report to the FDA in real time and still wait months for its documents to catch up.
The cost on the path to commercialization
Tufts CSDD puts the direct cost of a Phase III trial at $55,716 per day, in 2023 US dollars, based on budgets from 409 protocols. It remains the most recent Tufts benchmark. Applied across 6 to 20 weeks of startup delay, that is $2.3m to $7.8m for a single Phase III study. The same research puts Phase II at $23,737 per day, Phase II and III combined at about $40,000 per day, and unrealized prescription sales at about $500,000 for every day a product reaches market later.
Direct cost is only part of it. Every week spent preparing documents is a week later to data readout, a week later to submission, and a week less of commercial exclusivity once the product reaches market. For a biotech funding its pipeline milestone by milestone, startup time is runway.
The FDA is removing time from the review end of development. Sponsors who leave the startup end untouched will find it becomes the largest delay they still control.
What operational AI changes
Real time reporting accelerates a trial once it is running. Operational AI accelerates the part that comes first.
Clinials generates trial documents in minutes, directly from your protocol and supporting documents such as lab and pharmacy manuals.
One protocol upload produces:
Schedule of Activities, with procedures, biomarkers and endpoints organised by visit
Operational protocol synopsis for study teams
Visit worksheets and source documents for screening, baseline, treatment, follow up and unscheduled visits
Informed consent forms and plain language, multilingual patient materials
Budget discovery to support feasibility and early budget planning
Site initiation visit training decks
Every output is traceable to the protocol, version controlled and built for regulated workflows. Your team spends its time reviewing and approving, not rewriting. When a protocol is amended, materials are regenerated rather than rebuilt by hand.
No eClinical integration is required. Sponsors, CROs and sites work from the same protocol derived source, which removes the rework that happens when each party builds its own version.
Three questions for your clinical operations lead
How many weeks sit between final protocol and first site activation on our next study?
How much of that time is document creation, and rework after amendments?
What would it be worth to have those materials ready in minutes, not months?
If the answers are measured in months, startup is the most valuable time you can still recover.
Start faster
The direction of travel is clear: faster review, continuous trials and less tolerance for avoidable delay. The sponsors who benefit first will be those whose trials are ready to start when the protocol is final.
Your team can get started today
