Microsoft and Nvidia stand for it, and the British AI unicorn is regarded as "the hope of the whole village" by the prime minister

Author | Li Dongmei

The UK's self-driving unicorn received $1 billion in financing, with Microsoft and Nvidia as its platforms

Recently, Wayve, a British self-driving startup, announced that it has raised $1 billion in its latest round of funding, and after this financing, Wayve's total funding has exceeded $1.3 billion, making it the largest investment in the history of a British AI startup.

It is reported that the Series C financing was led by the Japanese conglomerate SoftBank, with AI giants Nvidia and Microsoft also participating.

Founded in 2017, London-based Wayve is one of many startups seeking to make autonomous driving a reality, with the company's technology allowing cars to drive effectively without humans at the helm.

Wayve also manufactures cars itself and licenses its self-driving technology to other companies, including retailers and automakers.

Founded by PhD students Alex Kendall and Amar Shah at the University of Cambridge, the company has raised $300 million in previous funding rounds. The company's valuation after the latest wave of venture capital has not been disclosed.

Wayve will use the funding to expand and release its own AI software product, so-called "Embodied AI," which means that such systems will work with any autonomous device that needs to navigate and act in a physical environment. Eventually, these systems will be deployed in the products of the major automotive manufacturing giants. The company says it's a novel approach to self-driving that allows vehicles to navigate better without following strict patterns or rules, such as when other human drivers act unexpectedly, or when a pedestrian runs onto the road without looking, or when a tree is suddenly blown onto the road.

Unlike previous generations of autonomous driving AI, the startup has developed software models that are trained "end-to-end" – including receiving camera and sensor data about what's happening around the vehicle, and then outputting the best possible driving action. The company has also trained an AI system that matches large language models with driving models so that the vehicle can interpret what it sees, why it is taking certain actions, and can accept natural language instructions.

Using AI models to make driving decisions, Wayve has found a bridge between AIGC and autonomous driving

Early self-driving cars relied heavily on multiple small AI models, each of which was responsible for performing a specific action, such as identifying objects in camera data, and then combining those small models with sophisticated, rule-based software to make driving decisions.

"This important funding milestone underscores our team's unwavering belief that Embodied AI will solve the long-standing challenges facing the autonomous driving industry in bringing this technology to the world," Wayve co-founder and CEO Kendall said in a statement Tuesday. ”

In addition, Alex Kendall said in an interview with Reuters: "This technology will allow automakers to accelerate the transition from assisted driving to autonomous driving."

Self-driving car startups have become more difficult to get funding than they have in the past few years, which is why Wayve's large funding has attracted so much attention.

In fact, many of today's autonomous vehicle developers are having some problems developing cars that are truly capable of driving themselves. This challenge is perceived to be much more difficult than many people think, so enthusiasm for autonomous driving has waned.

The autonomous driving industry has also experienced some setbacks recently.

Last October, a woman in San Francisco was reportedly towed underneath a car operated by Cruise, the autonomous vehicle arm of General Motors, after being hit by another human-driven vehicle. Cruise was then accused of concealing evidence about the accident, which led to it being deprived of its license to test vehicles on public roads.

At the end of the day, one of the biggest challenges with autonomous vehicle technology is that autonomous vehicle systems lack the ability of humans to predict and assess dangerous situations in real time, which is dangerous in the event of an unexpected event.

Wayve has been working hard to solve these headaches with its embedded AI system, which it claims is similar to "GPT for Driving," enabling any vehicle to better perceive its surroundings and drive safely in different environments. The software is designed to learn driving rules and patterns on its own without the need to program them, so it responds to new locations and unpredictable scenarios on the road.

Wayve President Erez Dogan told Reuters that Wayve's AI model is capable of generalizing its driving knowledge from one scenario to another. This is necessary, he says, because "it's almost impossible to imagine every situation that a self-driving car needs to handle reliably." ”

Dogan argues that AI can enable a paradigm shift in the way machines interact with and learn from human behavior. "By harnessing the raw power of AI, we can build an embedded AI system that learns from real-world and synthetic data how to handle edge cases faster than humans can program," he said.

A number of AI bigwigs stood up for it, and the British Prime Minister personally praised it

The investment is also a big win for the UK. The UK now has twice as many companies focusing on AI technology as other European countries, providing more than 50,000 jobs and contributing £3.7 billion ($4.6 billion) in economic benefits. British Prime Minister Rishi Sunak has been working hard to establish the country as a global AI hub, hoping to be at the forefront of technological development and regulation.

The support expressed by a number of leading companies with real money has also given the UK more confidence in this. In April, Microsoft announced that it would open an office in London dedicated to AI research.

It is worth mentioning that there are also a number of industry leaders who stand up for Wayve, including Yann Le Cunn, chief scientist at Microsoft and Meta.

Rishi Sunak also expressed his appreciation for the news of Wayve's latest funding round, saying it "proves our leadership in the industry."

"From the first light bulb or the World Wide Web, to artificial intelligence and self-driving cars, the UK has a proud track record at the forefront of some of the biggest technological advancements in history," Rishi Sunak said in a statement. ”

"I'm very proud that the UK is home to pioneers like Wayve who have made breakthroughs in developing the next generation of AI models for autonomous vehicles."

Between 2018 and 2022, the autonomous vehicle industry brought £475 million ($596 million) of direct investment to the UK, according to the UK government. It also created 1,500 new jobs, the government said.

With this round, Wayve has raised more than Germany's Aleph Alpa ($500 million in November) and France's Mistral ($415 million in December).

Commenting on Wayve's fundraising, Sunak said, "The fact that a UK-based company has set a record for the largest funding round for an AI company in the UK to date is a testament to our leadership in the industry. The news itself will also cement the UK's international position as an AI superpower. ”

In response to a request for comment, Wayve said it was reflected in the press release.

The obstacles will remain here for a long time

Major automakers such as Tesla and General Motors have already made moves in the field of autonomous driving, but the industry as a whole still faces challenges. Despite the ambitions of the world's giants to make driverless cars a reality, safety, economic costs and legal gaps remain significant long-standing obstacles.

Last year, Cruise, GM's self-driving car division, recalled vehicles for software updates while laying off a quarter of its employees. Tesla's "Full Self-Driving (FSD)" technology failure led the U.S. Department of Justice to investigate the company led by Elon Musk. The latest generation of Tesla's Full Self-Driving system uses an end-to-end approach similar to Wayve's.

Apple also recently canceled its long-awaited self-driving car project.

But things seem to be different on the Wayve side.

Its technical solutions are able to explain to the driver how the AI "thinks". They use real-world video and data from partners like Asda and Ocado to guide AI systems on how to navigate different road scenarios. Kendall said in an interview last September that the company's goal is to make self-driving car technology safer and more trustworthy among drivers. He also added that, overall, the odds of success continue to improve.

In an interview at the time, Kendall concluded, "I think the UK is a great fit for this technology. Based on what we have learned, the UK government also sees this as a huge growth opportunity and is willing to invest patience and time into it. ”

Original link: 7.5 billion yuan! This year's largest financing in the field of autonomous driving is coming: Microsoft, Nvidia for its platform, the British AI unicorn is regarded by the Prime Minister as "the hope of the whole village" _ Autonomous driving _ Li Dongmei _InfoQ selected articles