Tech
Hey, Waymo, have you tried the Fresh Pond rotary yet?
(Craig F. Walker/Getty Images)

To speed or not to speed? Tesla and Google’s Waymo disagree

Experts say going faster makes accidents more likely and potentially more harmful: “ Do we really want computers to make the decision to put other lives at risk because they want to break traffic rules?”

Google’s Waymo and Tesla are racing to win the driverless taxi market, but only Tesla appears to be going above the speed limit to get there.

A number of Tesla robotaxi videos show the vehicles going five or more miles per hour above the speed limit. Meanwhile, Waymo’s policy is to follow the posted speed limit, though the company says it will go below for construction areas or slightly above to change a lane, for example.

It’s an interesting difference as both companies try to win over the public’s trust in their new — and potentially dangerous — technologies. Tesla influencers who are part of the company’s limited launch have maintained that robotaxis go above the speed limit to match the speed of traffic, a practice they see as safer since it doesn’t disturb the flow of traffic. Tesla’s autonomous technology is trained on real-life drivers, and presumably it’s not screening out speeders in the training data.

The practice of speeding in self-driving Teslas lines up with what the company allows consumers to do in its supervised full self-driving cars. Tesla’s Model Y owner’s manual details a setting called “Max Speed Offset,” which reads:

Max Speed Offset: Set the percentage offset over the currently detected speed limit that Full Self-Driving (Supervised) can drive if it is necessary to drive faster than the speed limit to match the flow of traffic.”

Tesla didn’t respond to requests for comment about the incidents or its speeding policy.

Waymo recently collected data from its coverage areas in Phoenix and San Francisco and found that 33% to 49% of human drivers there were speeding, depending on the road type and location. The company contends that its speed-limit-following vehicles are safer than human drivers, and that speed compliance in those two cities alone could reduce traffic fatalities by 82 deaths annually. (Waymo hasn’t yet released data for its newer Austin market, where Tesla is also operating.)

“I would not be confident that they would see me and that they would detect me.”

Tesla CEO Elon Musk has repeatedly said that Tesla’s full self-driving will be significantly safer than human drivers.

“The standard has to be very high because the moment there’s any kind of accident with an autonomous car, that immediately gets worldwide headlines, even though about 40,000 people die every year in car accidents in the US, and most of them don’t even get mentioned anywhere,” Musk said on an earnings call this year. “But if somebody scrapes a shin with an autonomous car, it’s headline news.”

But when it comes to speeding, there are some hard truths.

Ken Kolosh, statistics manager at the National Safety Council, says that speeding makes accidents more likely and more dangerous when they do happen. The reasons are pure physics: when you’re traveling at a faster speed, it takes longer to stop, so it’s harder to avoid objects — or people — in your path. And higher speeds mean more damage and death.

“Nearly 3 in 10 traffic deaths in 2023 involved speeding — that’s 11,775 people killed, or more than 32 deaths every day,” he told Sherwood News.

The findings are the same over at the National Highway Traffic Safety Administration.

Phil Koopman, an associate professor of electrical and computer engineering at Carnegie Mellon University who specializes in autonomous vehicle safety, says the situation is a little complicated since the speed limit doesn’t always represent the appropriate speed, which varies by road type, condition, and weather, among other variables, but that generally the faster a vehicle drives, the more dangerous it is — even just five miles per hour faster.

Importantly, Koopman also says the comparison between robots and humans speeding doesn’t track because autonomous cars don’t have the same disincentives that people do, like traffic tickets or jail, if they violate the law or hurt someone.

“ Do we really want computers to make the decision to put other lives at risk because they want to break traffic rules?” he said.

When asked if he would personally ride in a Waymo or a Tesla robotaxi, Koopman said he would ride in the former but not the latter. He also said asking which is safer for the rider is the wrong question.

Instead, he pointed to other vehicles, cyclists, and pedestrians as where the concern should lie when considering autonomous vehicles. He said he personally would not walk in front of either.

“I would not be confident that they would see me and that they would detect me,” he said.

More Tech

See all Tech
tech

Elon Musk says Tesla Robotaxis are operating without drivers, sending stock higher

Tesla CEO Elon Musk said that Tesla’s Robotaxis are now operating in Austin without a safety monitor. Tesla has been testing driverless cars in the area for about a month, and Musk had previously said the company would remove safety drivers by the end of 2025.

It’s unclear how many exactly of the roughly 50 Robotaxis the company operates in the area don’t have drivers. Tesla is “starting with a few unsupervised vehicles mixed in with the broader robotaxi fleet with safety monitors, and the ratio will increase over time,” Ashok Elluswamy, Tesla’s head of AI, posted shortly after Musk. Ethan McKenna, the person behind Robotaxi Tracker, estimates it’s two or three vehicles.

What is clear is that the move is good for Tesla’s stock, which is currently up 3.5%, extending its gains after Musk’s tweet. Morgan Stanley said yesterday that it considers the removal of safety drivers a “precursor to personal unsupervised FSD rollout.” Unsupervised Full Self-Driving is widely considered to be integral to the would-be autonomous company’s value proposition.

At the World Economic Forum earlier on Thursday, Musk said, “Self-driving cars is essentially a solved problem at this point.”

tech

Survey: CEOs and workers have wildly different thoughts on AI productivity gains

One of the main reasons companies are rushing to adopt AI is to give their workers the miraculous productivity boost that AI companies have been promising — and believe will quickly earn back their investment.

But now that companies have been using AI for a while, a growing perception gap is emerging between the C-suite and their employees.

The Wall Street Journal reported on new findings by research firm Section, which surveyed 5,000 white-collar workers from companies with more than 1,000 employees.

More than 70% of the corporate executives in the survey said they were “excited” by AI, and 19% of them said the tools have saved them more than 12 hours of work per week.

But nonmanagement workers had a very different take on AI. Almost 70% of this group said AI made them feel “anxious or overwhelmed,” and 40% said the tools saved them no time at all.

The Wall Street Journal reported on new findings by research firm Section, which surveyed 5,000 white-collar workers from companies with more than 1,000 employees.

More than 70% of the corporate executives in the survey said they were “excited” by AI, and 19% of them said the tools have saved them more than 12 hours of work per week.

But nonmanagement workers had a very different take on AI. Almost 70% of this group said AI made them feel “anxious or overwhelmed,” and 40% said the tools saved them no time at all.

tech

Tesla jumps as Musk says he expects Optimus sales next year, European and Chinese FSD approval next month

Tesla CEO Elon Musk now says he thinks the company’s Optimus robots will be for sale to the public “by the end of next year.”

According to Musk, “That’s when we are confident that there is very high reliability, very high safety, and the range of functionality is also very high.”

Like many of Musk’s other timelines, that’s later than he previously predicted. In 2024, for example, Musk said the AI robots would be for sale in 2025.

Speaking with BlackRock CEO Larry Fink on a panel today at the World Economic Forum, Musk said the robots are currently doing “simple tasks” in Tesla factories, but believes “they’ll be doing more complex tasks and be deployed in an industrial environment” by the end of this year, before going on sale to the public in 2027.

Musk forecasts a future with “billions” of AI robots that “saturate all human needs.”

On a separate topic, Musk was bullish on regulatory approval for what Tesla calls Full Self-Driving technology in markets outside the US. “We hope to get supervised Full Self-Driving approval in Europe, hopefully next month, and then maybe a similar timing for China,” he said. Musk has said in the past that the pending regulatory approval for FSD in Europe is a key reason why Tesla’s sales in the region have been tanking.

tech

Waymo is now offering autonomous rides in Miami

Google subsidiary Waymo announced Thursday that it’s officially open for autonomous ride-hailing in Miami, expanding the company’s coverage area to six US cities. The company will be “inviting new riders on a rolling basis” to take rides across its 60-square-mile service area, which includes the Design District, Wynwood, Brickell, and Coral Gables. Waymo said it plans to expand to Miami International Airport “soon.”

Competitor Tesla currently operates a ride-hailing service with a safety monitor in the vehicle in Austin and the Bay Area.

tech

Apple to promote Siri from assistant to chatbot

Bloomberg reports that Apple plans to transform its Siri assistant into a full-fledged chatbot similar to OpenAI’s ChatGPT.

The chatbot would be integrated throughout the iPhone’s operating system rather than offered as a stand-alone app. It’s expected to arrive later this year and would be separate from more incremental, non-chatbot improvements to Siri rolling out in the coming months aimed at making the existing assistant more usable.

Both updates will be powered by Google’s AI models, Bloomberg reports, but the chatbot upgrade will be more advanced and akin to the much-lauded Gemini 3.

While the difference between an assistant and a chatbot may sound subtle, it represents a meaningful shift for Apple, which has long avoided a fully conversational interface and has lagged rivals that embraced one. Any new Siri chat capabilities could also eventually extend to other Apple devices under development, including wearables such as the pin Apple is developing.

Both updates will be powered by Google’s AI models, Bloomberg reports, but the chatbot upgrade will be more advanced and akin to the much-lauded Gemini 3.

While the difference between an assistant and a chatbot may sound subtle, it represents a meaningful shift for Apple, which has long avoided a fully conversational interface and has lagged rivals that embraced one. Any new Siri chat capabilities could also eventually extend to other Apple devices under development, including wearables such as the pin Apple is developing.

Latest Stories

Sherwood Media, LLC produces fresh and unique perspectives on topical financial news and is a fully owned subsidiary of Robinhood Markets, Inc., and any views expressed here do not necessarily reflect the views of any other Robinhood affiliate, including Robinhood Markets, Inc., Robinhood Financial LLC, Robinhood Securities, LLC, Robinhood Crypto, LLC, or Robinhood Money, LLC.