Updated: September 8, 2026
A car can advertise automatic emergency braking, lane keeping, blind-spot intervention and even hands-free highway assistance — yet those features are not all tested in the same way, and they do not roll into one universal “ADAS safety score.”
That is where many buyers get confused. In the United States, NHTSA and the Insurance Institute for Highway Safety (IIHS) both evaluate vehicle safety technology, but they ask different questions and publish different kinds of results.
Understanding that difference matters more in 2026 because NHTSA has expanded its New Car Assessment Program with new ADAS performance tests, while IIHS has continued to raise the bar for crash prevention and driver-monitoring safeguards.
Do Not Treat Every Safety Rating as the Same Rating.
NHTSA tests whether specific driver-assistance technologies meet defined performance criteria. IIHS separately evaluates crash prevention, overall vehicle safety and the safeguards built into partial driving automation. A strong result in one category does not automatically guarantee a strong result in another.
IIHS: independent ratings for crashworthiness and crash avoidance, plus a separate safeguard rating for Level 2 partial automation.
Buyer takeaway: check crash protection, crash-prevention performance and automation safeguards separately.
Start here if you want a plain-English explanation of AEB, adaptive cruise control, lane keeping, lane centering and SAE automation levels.
First: NHTSA's 5 Stars Are Not an “ADAS Score”
NHTSA's New Car Assessment Program, or NCAP, is best known for the 1-to-5-star safety rating displayed for new vehicles. Those stars primarily communicate crash protection and rollover safety.
Driver-assistance technology is evaluated alongside NCAP, but the result should not be interpreted as “this car has five-star ADAS.” NHTSA also identifies driver-assistance technologies that have met its performance tests, and in 2026 the agency began highlighting a broader pass/fail ADAS benchmark.
What Changed in NHTSA ADAS Testing for 2026?
NHTSA expanded NCAP by adding four advanced crash-avoidance technologies and strengthening evaluations of technologies already in the program.
Those four were added to the four ADAS areas NHTSA was already evaluating:
- Forward Collision Warning (FCW)
- Crash Imminent Braking (CIB)
- Dynamic Brake Support (DBS)
- Lane Departure Warning (LDW)
The 2026 Tesla Model Y Milestone
On May 7, 2026, NHTSA announced that the later-release 2026 Tesla Model Y became the first vehicle model to pass the agency's new expanded ADAS benchmark.
NHTSA specified that the result applies to 2026 Model Y vehicles manufactured on or after November 12, 2025. The vehicle met the agency's criteria for the four newly integrated tests — pedestrian AEB, lane keeping assistance, blind spot warning and blind spot intervention — and also passed the four original ADAS criteria.
NHTSA vs IIHS: The Simplest Comparison
| Question | NHTSA | IIHS |
|---|---|---|
| Who runs it? | U.S. federal safety regulator | Independent nonprofit safety research organization |
| Familiar overall result | 1–5 Star Safety Rating | Good / Acceptable / Marginal / Poor test ratings; Top Safety Pick awards |
| ADAS crash avoidance | Performance testing and new pass/fail benchmark | Front crash prevention tests for vehicle and pedestrian scenarios |
| Partial automation | Driver assistance remains supervised; not represented by a self-driving score | Separate safeguard rating for Level 2 systems |
| Best use for buyers | Check crash protection plus whether key technologies meet federal test criteria | Compare crashworthiness, crash prevention and automation safeguards in more detail |
How IIHS Tests Vehicle-to-Vehicle Crash Prevention
IIHS's current vehicle-to-vehicle front crash prevention evaluation is much more demanding than a simple “does the car have AEB?” checklist.
The test vehicle approaches stationary targets representing a passenger car and a motorcycle. Both warning and automatic braking are evaluated. IIHS also uses an actual dry-van semitrailer for forward-collision-warning testing.
Current testing includes speeds of 31, 37 and 43 mph. Passenger-car and motorcycle targets can be positioned in the center of the lane or offset to one side, reflecting situations that can be more difficult for sensors and software to interpret.
IIHS currently uses the familiar four-grade scale:
Good Acceptable Marginal Poor
How IIHS Tests Pedestrian AEB
Pedestrian AEB is especially important because detecting a person can be harder than detecting a vehicle, particularly at night.
IIHS evaluates several common crash scenarios:
This is one reason two vehicles that both advertise “pedestrian detection” can perform very differently in independent testing.
What IIHS Partial Automation Safeguard Ratings Actually Measure
This is the category buyers most often confuse with crash-prevention ratings.
IIHS's partial automation safeguard rating does not primarily answer “How well does the car steer itself?” Instead, it asks whether the system is designed to keep the human driver engaged while Level 2 assistance is active.
IIHS evaluates areas including:
- Driver monitoring
- Attention reminders and warning escalation
- Emergency procedures if the driver does not respond
- Automated lane-change safeguards
- Adaptive-cruise-control restart behavior
- Cooperative steering
- Whether key safety systems remain active
Apply these rating concepts to three of the best-known U.S. Level 2 systems and compare their operating design, road coverage and driver-monitoring approach.
IIHS currently says partial driving automation should be viewed as a convenience feature, not a proven stand-alone safety system. The institute says real-world evidence has not established a convincing crash-reduction benefit beyond crash-avoidance technologies such as AEB and lane-departure prevention.
See why hands-free does not mean driver-free, where Level 2 systems can struggle and why driver monitoring matters on long highway trips.
What Counts for IIHS Top Safety Pick in 2026?
IIHS also uses crash-prevention performance as part of its annual vehicle awards, but again, this is separate from the partial-automation safeguard rating.
For 2026 Top Safety Pick+, IIHS requires:
- Good small-overlap front crash-test performance
- Good moderate-overlap front crash-test performance
- Good side-crash performance
- Acceptable or good headlights as standard
- Good pedestrian front crash prevention as standard
- Acceptable or good vehicle-to-vehicle front crash prevention 2.0 as standard
The 2026 Top Safety Pick base award still requires acceptable or good pedestrian front crash prevention, but the stronger vehicle-to-vehicle 2.0 requirement is part of the “Plus” standard.
Three Ratings Buyers Commonly Mix Up
| Result | What It Mainly Tells You | What It Does Not Tell You |
|---|---|---|
| NHTSA 5-Star Rating | Crash protection and rollover safety | Complete performance of every ADAS feature |
| IIHS Front Crash Prevention | How warning/AEB performs in defined vehicle or pedestrian scenarios | How well a Level 2 system keeps the driver attentive |
| IIHS Partial Automation Safeguards | Driver monitoring, attention reminders and misuse safeguards | Overall crashworthiness or full obstacle-detection performance |
Why “My Car Has ADAS” Is Not Enough
Feature availability tells you only that the hardware and software exist. It does not tell you how well the feature performs.
Two cars can both list automatic emergency braking, for example, but one may detect a motorcycle or pedestrian more reliably across difficult test scenarios. Two cars may both offer lane centering, but one may have much stronger driver-monitoring and warning escalation.
That is why independent and government testing is valuable: it moves the comparison away from marketing names and toward measurable performance.
How to Read ADAS Ratings When Shopping for a Car
Does a High ADAS Rating Mean You Can Trust the Car More?
It means you have stronger evidence that the tested system performed well under the specific test conditions. That is valuable.
But no controlled test can reproduce every road, weather, lighting, construction-zone, motorcycle, pedestrian or driver-behavior scenario you will encounter.
A strong result should therefore increase your confidence in the tested capability, not encourage you to surrender attention.
Will ADAS Ratings Become More Important?
Almost certainly.
NHTSA's expanded NCAP testing is part of a longer-term roadmap to keep adding and strengthening crash-avoidance evaluations. IIHS has also repeatedly raised the difficulty of its front crash-prevention and award criteria as vehicle technology improves.
That creates a healthy cycle: once most vehicles perform well on an older test, the test becomes harder, pushing automakers to improve performance in more difficult and more realistic scenarios.
For buyers, this means a simple feature checklist will become less useful. How well the system performs will matter increasingly more than whether the brochure says the car has it.
DriveSpecLab Verdict
The biggest mistake in ADAS shopping is looking for one magic safety score.
NHTSA and IIHS are measuring different layers of the safety problem. NHTSA tells you about crash protection and whether specific driver-assistance technologies meet government performance criteria. IIHS goes deeper into crash prevention and also asks whether Level 2 systems have strong safeguards to keep the driver engaged.
The best car is not simply the one with the longest ADAS feature list. It is the one that combines strong crashworthiness, proven crash-avoidance performance and sensible driver-monitoring safeguards.
And even then, ADAS remains exactly what the name says: assistance.
Frequently Asked Questions
Does NHTSA give cars an ADAS star rating?
No. NHTSA's familiar 1-to-5-star rating summarizes crash protection and rollover safety. Driver-assistance technologies are evaluated through separate performance testing, including the expanded ADAS pass/fail benchmark introduced within NCAP.
What was the first vehicle to pass NHTSA's new 2026 ADAS benchmark?
NHTSA announced on May 7, 2026 that later-release 2026 Tesla Model Y vehicles manufactured on or after November 12, 2025 were the first model configuration to pass the expanded benchmark.
What four new ADAS technologies did NHTSA add to NCAP?
Pedestrian automatic emergency braking, lane keeping assistance, blind spot warning and blind spot intervention.
Does an IIHS Top Safety Pick+ mean the Level 2 system is good?
Not necessarily. Top Safety Pick+ considers crashworthiness, headlights and specified crash-prevention performance. IIHS publishes a separate partial-automation safeguard rating for Level 2 systems.
What does the IIHS partial automation rating measure?
It focuses on safeguards such as driver monitoring, attention reminders, emergency procedures, automated lane-change behavior, cooperative steering and whether key safety systems remain active.
Can a car have good AEB but weak partial-automation safeguards?
Yes. Crash-avoidance performance and partial-automation safeguards are different evaluations. A vehicle can perform strongly in one category and less strongly in another.
Should I choose a car based only on ADAS ratings?
No. Use ADAS results alongside crashworthiness, headlights, braking, tires, visibility, reliability, ergonomics and the exact trim-level equipment you are buying.
Official Sources
- NHTSA — 2026 Tesla Model Y First to Pass New ADAS Tests
- NHTSA — 5-Star Safety Ratings & Driver Assistance Technologies
- NHTSA — Driver Assistance Technologies
- IIHS — 2026 Top Safety Pick Criteria
- IIHS — Vehicle-to-Vehicle Front Crash Prevention Test
- IIHS — Pedestrian Front Crash Prevention Test
- IIHS — Partial Automation Safeguard Ratings
DriveSpecLab editorial note: Safety ratings can apply to specific model years, trims, production dates, hardware or software configurations. Always verify the exact vehicle you are considering and consult the manufacturer's owner's manual for feature limitations and operating conditions.
