Cybersecurity breaches are about to get worse. Much worse.

For all the attention we’re paying to what AI will help us build, it seems we need to spend more time considering what AI will help attackers break.

Cybersecurity breaches will quickly become more frequent, more disruptive, and harder to contain. The danger lies in a growing mismatch: AI is accelerating the discovery and exploitation of vulnerabilities, while securing the organizations exposed to them remains a complicated, often slow process.

This week provided a warning. AI-assisted research helped uncover two critical flaws in PHP, software used across much of the web. OpenAI and its partners spent roughly two months coordinating a response. Yet Amazon subsequently identified an exploitation attempt four days before the vulnerabilities were publicly disclosed.

Finding the flaws was only the beginning. Getting protection into every affected system was the harder problem.

(Click here to continue reading)

Until next week,

Steve Greenfield
General Partner
Automotive Ventures

🚗 Automotive

Financial distress continues to affect trucking companies across the U.S., with at least 16 carriers filing for bankruptcy protection between Sept. 18 and Oct. 21. According to a FreightWaves review of federal bankruptcy records, the filings span multiple freight segments, including general freight, petroleum transportation, logging and timber hauling. They also involve companies ranging from single-truck operators to carriers reporting dozens of power units. | FreightWaves

American roads are becoming deadlier for walkers. Between 2009 and 2024, pedestrian deaths increased 72% across the U.S., federal data show. Over the same period, such deaths declined in most other developed nations, according to United Nations data. In road safety, the U.S. is an outlier. | The Wall Street Journal ($)

Pirelli's new "Cyber Tire" has a Bluetooth sensor slightly larger than an Apple AirTag and shaped like a starfish. It lives inside the tire and can withstand a tremendous amount of stress as it goes round and round. A receiver module mounted on the axles transmits all gathered information to the car’s onboard computer. Specifically, it collects data on tire pressure, temperature, grip, and even wear. It then communicates with the car via an algorithm that adjusts braking (ABS) and stability (ESP) thresholds and application in real time. In a recent test performed by evo, the sensor-driven tire delivered shorter stopping distances, better grip, and impressive data collection compared to non-Cyber tires, giving drivers and automakers alike valuable information to improve performance and safety. | The Drive

🇨🇳 China

In the span of a few years, China and the United States have each assembled something future historians might recognize as a new kind of empire. Their dominions extend not merely over territory, trade routes, and airspace but over knowledge itself. By controlling the large language models, chips, physical infrastructure, and capital that provide the foundations for advanced machine cognition, Beijing and Washington are not only attaining new heights of economic performance and power projection; they are also setting the terms of all future scientific and intellectual progress. The term “AI superpowers” fails to capture the influence and reach that these rivals now wield. Even without active alliance building or coercion, they can make all other countries that seek access to advanced AI beholden to them. Imperial history carries a warning. In the absence of other forces, empires seldom succumb to a challenger or to barbarians at the gates. More often, they fail because defending their vast dominions drains them dry, whether of material resources, the consent of the governed, or the loyalty of elites. The imperial center finds it can no longer maintain control over the periphery—the regions that once submitted to it. Failure can be financial, as it was for the British Empire after 1945, when the debts of two world wars forced London to liquidate its imperial assets to pay the price of victory: a welfare state for the citizens of the imperial metropole. It can be institutional, as it was for the Qing empire in the late nineteenth century, when it failed to industrialize, reform its bureaucracy, and renew the legitimacy of the Manchu dynasty. Or it can be technological. The Royal Navy’s Dreadnought of 1906 gave the United Kingdom the first “all big gun” battleship and, with the same stroke, rendered its existing battle fleet obsolete. Overnight, this breakthrough opened the way for rivals such as Germany, erasing a century of accumulated British advantage. The resulting arms race forced the United Kingdom to nearly double naval spending between 1900 and 1914 and clashed with the Liberal government’s new social spending commitments, such as the Old Age Pension Act of 1908. The decisive question in the contest between Beijing and Washington may therefore not be who achieves artificial general intelligence first, but who can bankroll the race. | Foreign Affairs ($)

Chery is open to partnerships with U.S. automakers, a senior executive told Reuters on Wednesday, as China's top car exporter looks to expand in North ‌America following rapid growth in Europe and elsewhere. For now, Chinese car companies remain effectively locked out of the United States, the world's second-largest auto market, where they face tariffs of more than 100% and data security regulations. "The U.S., obviously, we are trying to look for different options there," Zhang said, emphasizing Chery's flexibility. "We're very open to different options, even a potential partnership with American automakers," he said. Zhang's comments are some of the clearest yet about Chery's ‌American ⁠ambitions, although he did not go into specifics about the kind of partnerships he envisions. Zhang said he was confident that Chery would find a way to target the U.S., saying it would be possible to build a car that was not reliant on Chinese hardware or ⁠software. "Definitely from a technical point of view, we will be able to find a China-free solution in terms of hardware and software," he said. | Reuters ($)

Pressure on automakers has been building on all sides. Domestically, Tesla has been pushing the boundaries of autonomy, in part because it developed a true software-defined vehicle. Automakers like General Motors are also feeling pressure from Chinese manufacturers, which have shortened development cycles to two to three years, down from the usual four to five years. GM intends to match that cadence. Alongside announcements about battery chemistries and autonomy, GM recently outlined how enterprise AI tools have sped up vehicle development across every sector as it pushes to slash its traditional four-to-five-year development timeline in half. | TechCrunch ($)

🤖 Autonomy

Tesla said Wednesday it will start selling its Cybercab to customers by the end of the year, even as federal regulators investigate whether the vehicle meets vehicle safety standards. The gold-hued, autonomous two-seater lacks a steering wheel and other traditional controls. During the company’s third-quarter earnings call, Lars Moravy, vice president of engineering at Tesla, said the Cybercabs produced today are all destined for the company’s own robotaxi fleet. As that fleet grows, the company will “mainly deliver fully designed autonomous vehicles to our customers,” he said, noting that it will start small. The company will scale as it learns about delivery, ownership, and fleet management of autonomous vehicles, Moravy added. Tesla first floated the idea of selling the autonomous vehicle to third parties last month around its Cybercab event in Austin and even posted a form for people interested in buying robotaxis. But it wasn’t clear if Tesla would follow through, or how. | TechCrunch ($)

Tesla is returning to growth mode after years of “things being fairly flat,” according to CEO Elon Musk. The demand driver is Tesla’s Full Self-Driving software, Musk said on an Oct. 21 call to discuss the company’s third-quarter financial results. FSD requires driver supervision on personal Teslas but now operates autonomously on the company’s robotaxis. “Today, there’s really no competitor for FSD in any car that you can buy,” Musk said. The electric vehicle maker is expanding its factories with a production target of 5 million vehicles annually by 2030, Musk said. “We’ve got a good shot of achieving that goal,” Musk said. “After many years of sort of a grind and things being fairly flat, I think we’re now headed for a lot of fun and significant growth.” Tesla’s global sales peaked in 2023 at 1.8 million before edging lower a year later. Last year, deliveries and production both fell to just over 1.6 million, but they are trending higher this year. | Automotive News ($)

Tesla says it’s building its own maps to fix some of the most common problems with its “Full Self-Driving” system, including reading speed limits, parking, and choosing routes. The company’s head of AI described the effort on Tesla’s Q3 2026 earnings call. It’s a notable shift for a company whose CEO spent years arguing that self-driving systems relying on maps are brittle. And it shows Tesla’s fleet is now doing the same job Waymo’s cars do: mapping the roads so the self-driving system knows what it can’t reliably figure out on its own. Tesla fans argue that Waymo’s approach differs because it creates HD maps with lidar, but ultimately both are creating their own maps; if your argument is that Tesla’s maps are less well-defined and needed for driving for different reasons than Waymo’s, you don’t really have an argument. | Electrek

A top candidate to become General Motors’s next chief executive said the company’s future lies in autonomous vehicles that will help it leapfrog global rivals, including from China. In an interview with the Financial Times, Sterling Anderson, a former Tesla executive charged with overseeing GM’s next generation of electric and autonomous vehicles, outlined a vision for autonomous vehicles that can ferry people to work but also pick up their shopping for them when they are at the office. He said he believed people would still prefer to own their own cars rather than rely on a robotaxi service. | Financial Times ($)

There are many things humans do that are difficult to imagine any machine improving upon, but driving is not one of them. Human drivers possess abundant and obvious shortcomings. A decade ago, Elon Musk told reporters that, if their skepticism turned the public against autonomous vehicles, they would be “killing people.” Waymo offers a more measured version of the same general argument: cars don’t kill people; people kill people. | The New Yorker ($)

The first state to pass a law legalizing self-driving cars operating on public roads was Nevada in June 2011 — long before such vehicles were operating on public roads. This was three years before the introduction of Tesla’s Autopilot, five years before Waymo spun out of the Google self-driving car project, and seven years before the first commercial robotaxi service got underway (in Arizona, not Nevada). Nevada was ahead of the curve primarily due to lobbying: Google saw Nevada as prime testing ground for its prototypes, so it urged lawmakers to pass the law that allowed it to do so. The legislation even included a provision exempting drivers from text-messaging bans while the automated system was engaged. Back then, many states and cities welcomed self-driving cars with open arms, eager to be seen as future-forward and pro-innovation. They had no inkling that over a decade later, they would be back at the drafting table crafting legislation to rein in these same self-driving cars. While federal regulators scramble to establish safety standards, legislation that would establish rules for autonomous vehicles remains stalled in Congress — and has been for years. Washington’s paralysis has left local lawmakers to create their own playbooks to ensure robot cars at least follow some rules of the road. The result is a patchwork of rules and regulations for robotaxi operators to follow — an outcome that many industry leaders were desperate to avoid. | The Verge ($)

U.S. auto safety regulators proposed a new category of cyan exterior lamps to identify driverless vehicles to other road users. One or two of the green-blue lights would be placed on the front, sides and rear of vehicles equipped with automated driving systems, at least 15 inches above the ground, according to the National Highway Traffic Safety Administration (NHTSA) proposal made public on Friday. The new lights, which would be optional, would mark a shift from the limited color palette of passenger-vehicle exterior lights, which for decades have been confined to red brake signals, amber turn indicators and white headlamps. Adding a new category of cyan exterior lights aims to help law enforcement and other road users distinguish self-driving cars from vehicles operated by humans. It’s part of a broader push by NHTSA to remove unintended barriers to autonomous vehicles posed by U.S. auto safety standards, many of which were developed long before the advent of self-driving technology. | Bloomberg ($)

🤖 Artificial Intelligence (AI)

General Motors said Wednesday that it was starting to use artificial intelligence to halve the time it took to design a new vehicle to as little as two years. If successful, the initiative would improve GM’s chances of competing with fast-moving Chinese automakers. BYD, Geely, Xiaomi and other Chinese manufacturers have stunned established carmakers by quickly rolling out new models packed with the latest technology. Sales of relatively inexpensive Chinese vehicles have soared in Asia, Africa, Europe and Latin America, often at the expense of U.S., European or Japanese brands. GM also said it would overhaul the computer systems in its cars and drastically cut the cost of electric vehicle batteries, which could also be seen as a response to Chinese rivals. Sterling Anderson, GM’s chief product officer, portrayed the changes as an effort to make better cars. “The faster we can iterate, the better our products become,” he said in an interview with The New York Times. “We are developing better products faster at lower costs.” | The New York Times ($)

Salesforce is leading the software industry’s attempt to collect fees from businesses that use Claude AI and other external AI tools to automate work in enterprise apps. As soon as next month, Salesforce plans to require companies to register outside AI agents, a step toward charging them each time those agents update or analyze data stored in Salesforce. The fees let Salesforce earn money from customers who use such AI instead of paying for Salesforce’s built-in AI products. The Salesforce tollgate has implications for the rest of the industry, which is coming to terms with the idea that so-called AI agents may be able to use software firms’ apps more efficiently than humans do, meaning their seat-based or per-user pricing model might not have staying power. The question is whether the new AI tollgates—which Atlassian, Workday, ServiceNow and Datadog are also erecting to various extents—and other pricing shifts will more than make up for an eventual decline in per-seat subscription revenues. | The Information ($)

Over the past couple of weeks, a cybersecurity researcher watched as a Chinese-speaking hacker broke into websites belonging to more than 100 companies, including farm equipment maker John Deere, home appliance company SharkNinja, and engine manufacturer Cummins Inc. Yet the hacker did almost none of the work. The cybercriminal instead outsourced the hacking spree almost entirely to artificial intelligence, tasking AI agents ‌with spending hours autonomously probing and breaking into targeted sites, according to Snorre Fagerland, principal threat researcher at the cybersecurity company NorForge Inc. In a report provided to Reuters, Fagerland said the hacker was one of several he discovered while scanning the web for inadvertently exposed instances of Cairn, an open-source AI problem-solving platform popular with hackers. Beginning around October 2, Fagerland said he found about two dozen instances of Cairn being used for malicious activity, which allowed him to quietly peer over the cybercriminals’ shoulders as they set their AI tools loose on the world. | Reuters ($)

Mass cybersecurity disruptions, such as this week’s scramble by governments and companies to patch two critical software flaws, are about to “get a lot worse before they get better,” a top executive at Anthropic has warned. Banks, airlines and governments around the world pulled down their websites on a precautionary basis this week to prevent them from being hacked after cybersecurity researchers, with the aid of AI, discovered vulnerabilities in the open-source programming language PHP that underpins huge swaths of the internet. "As bad as things have gotten in the past 4 days, they are going to get a lot worse before they get better,” Jason D. Clinton, deputy chief information security officer for Anthropic, wrote in a post on X on Thursday. “Security leaders, never let a crisis go to waste: please go to your boards and ask for help right now. Not for AI, for the basics.” AI models are increasingly able to uncover serious cybersecurity vulnerabilities — a capability that helped researchers find the two PHP flaws that led to the proactive service suspensions this week, but could also turbocharge hackers’ ability to exploit weaknesses. That has set a blistering new pace for cybersecurity, collapsing the time that organizations have to respond to AI-discovered threats to close to zero. | Bloomberg ($)

✈️ Aviation & Space

Elon Musk has engineered ties between Tesla and SpaceX that are hard to break, and now he wants to fuse them together permanently. Musk has run both Tesla and SpaceX concurrently over the past two decades, using a schedule that dictated which days he spent focused on each business. The companies have often shared resources and personnel, signaling their close ties. Last year, SpaceX looked into creating a spacesuit for Tesla’s humanoid robot Optimus, a person familiar with the matter said. Tesla last week showed off its new Roadster, a vehicle featuring SpaceX rocket thrusters that it hopes to soon bring to market. And in recent months, Musk has repositioned both companies by expanding beyond their core operations toward a shared mission centered on artificial intelligence. The world’s richest man on Tuesday signaled his interest in tying up SpaceX and Tesla, marking what would be the biggest corporate merger of all time. Because he serves as chief executive of SpaceX and Tesla, Musk said committees of independent directors at both companies are evaluating a possible deal, adding that a transaction might not come to fruition. | The Wall Street Journal ($)

With President Trump approaching the second half of his final term and jittery markets threatening the exuberance underpinning Elon Musk’s empire, there’s a push to announce a deal between SpaceX and Tesla by the first half of 2027 so that it closes during this administration, according to people familiar with the matter. That would bring together companies with a combined market value of around $3.7 trillion. It would also consolidate Musk’s corporate power before a potentially less-accommodating political and financial environment takes hold. The midterm elections are less than two weeks away. Democratic control of either chamber of Congress could expose Musk and his businesses to investigations and weaken a president already heading toward lame-duck status. A sustained selloff in chipmaker and artificial intelligence stocks, meanwhile, could erode the investor support needed to pull off such an audacious transaction. | Bloomberg ($)

At first glance, their already tight kinship, and sundry overlapping businesses, suggest that a SpaceX-Tesla makes sense. SpaceX buys Tesla Megapack batteries and Cybertrucks, Tesla invested $2 billion in xAI, bought by SpaceX in February, and they share ownership of the much-feted chip project Terafab. Not to mention that the deal would clinch Musk’s long-harbored vision of running a single colossus as a kind of Berkshire Hathaway of AI-driven tech. The wizard freed to blend these two great elixirs of innovation? What’s not to like? A close examination of claims made for the deal, especially assumptions on its supposed advantages in bolstering the finances of both companies, points to an extremely risky wager that must perform never-before-seen heroics to reward shareholders. It’s the ultimate buy the leap of faith, it’s about the vision, not the numbers, never bet against Elon scenario, set against the sobering numbers. | Fortune ($)

Elon Musk ordered the slowdown of a military-grade version of Starlink relied on by Ukraine, a decision that sparked panic in Kyiv and confusion inside the Trump administration, U.S. officials and people familiar with the matter said. Both Musk and President Trump have expressed anger at Ukraine for attacking Russian diesel refineries in recent weeks, which the president blames for high energy prices. Last week, Musk capped the speeds at which Ukrainian hardware could stay connected to the satellite-based network, the people said, rendering some offensive drones and interceptor drones virtually useless. Musk and Trump discussed the Starlink decision in a call soon after the slowdown became public, while the president was campaigning last week, two of the people said. One of the people also claimed Musk and Trump discussed the matter prior to the decision to throttle Starlink. | The Wall Street Journal ($)

After SpaceX restricted use of its Starkink and Starshield satellite internet services in Ukraine, causing fast-moving drones to disconnect and drop out of the sky, Kyiv is looking for alternatives. Ukraine’s Ministry of Defense said on Thursday that it has signed a letter of intent to develop new defense and satellite communication capabilities in partnership with Finnish tech companies Nokia and ICEYE, which makes satellites. “Modern warfare has proven that reliable communications are a matter of national security,” Ukraine President Volodymyr Zelenskyy said in a statement. “We must all rely on capabilities that we can control.” Zelenskyy also announced a partnership with Norway this week to jointly develop air defense and communications systems that could serve as an alternative to Starshield. | CNBC

🚘 Car of the Week

📢 Our Automotive Ventures “Car of the Week”: a 1955 Mercedes-Benz 300SL Gullwing. | Bring a Trailer ($)

🗞️ In the News

📢 Automotive Ventures portfolio company Go Eve has secured £750K in funding. | Go Eve

👀 Automotive Ventures Company to Watch

PRZM provides the data and workflow layer powering the auto salvage economy: Towers move vehicles faster, insurers see it in real time, and every auction sale pays out, guaranteed. | PRZM