What happened
Consumer Reports published its first ranking of partially automated driving systems on 4 October 2018. It tested four systems on its 327-acre track and on nearby Connecticut freeways, as they worked in September 2018. Tesla's Autopilot was tested on the Model X, Model S and Model 3. It scored 4 of 5 for capability and performance, the highest of the four. Consumer Reports found that Autopilot and Cadillac's Super Cruise accelerated and slowed comfortably. They reliably kept the car centered in its lane for several miles at a time. Testers found that Autopilot was easy to engage and easiest to use in stop-and-go traffic. It ranked second, behind Super Cruise. It scored 1 of 5 for keeping the driver engaged and waited 24 seconds before warning an inattentive driver. It also let drivers use it on small, curvy roads with poor lane markings, where it operated erratically. The tests measured system behavior and no crash outcomes. Consumer Reports writes that such systems use cameras, radar and other sensors but does not describe Autopilot's software.
Sources
-
Assessment report, by authors with no tie to the system.
Patrick Olsen, Consumer Reports. Cadillac Tops Tesla in Consumer Reports' First Ranking of Automated Driving Systems. 4 October 2018, ranking table, Autopilot entry. Ratings from tests by Consumer Reports of four systems on its track and on Connecticut freeways, September 2018.
Autopilot performed well and is easiest to use in stop-and-go traffic.
-
Press release, by the makers. Context or contrary evidence.
Tesla. All Tesla Cars Being Produced Now Have Full Self-Driving Hardware. Tesla blog, 19 October 2016, archived copy of 18 October 2020. Cited for the neural net that Tesla's onboard computer runs for vision processing.
To make sense of all of this data, a new onboard computer with more than 40 times the computing power of the previous generation runs the new Tesla-developed neural net for vision, sonar and radar processing software.
Details
AI system
- Workflow
-
Consumer Reports ran the same tests on all four systems on its track and on Connecticut freeways. Testers drove the cars several times over the same route, both as the leading and as the following vehicle. The evaluation covered the systems as they worked in September 2018.
Follow-up
- Conditions for replication
-
The driver stays engaged, since Autopilot waited 24 seconds before warning the driver to respond. The road suits the system, since Autopilot operated erratically on small, curvy roads with poor lane markings. Over-the-air updates may change how the system behaves after the tests.
Entry details
- Sector
-
Road vehicle crash avoidance / assisted driving.
- Purpose
-
Tesla built Autopilot to hold a set speed and following distance and to steer the car within its lane while the driver supervises.
31 more questions not yet checked
model, release, provider, role, configuration, inputs and outputs, autonomy, human oversight, AI contribution, beneficiaries, reach, distribution and access, duration, every measurement question, every money question, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.
Cite this record
Peter H. Vartanian. (2026). Autopilot kept Teslas centered for miles in CR's first ranking (AIBD-1105). Artificial Intelligence Benefits Database, edition of 1 November 2026. https://petervartanian.xyz/aibd/records/AIBD-1105.html (archived at https://doi.org/10.7910/DVN/AADSII, version 7.0)