The world is an incredibly complicated place, and we’ve only just started to accept the concept of photoshopped images, and now AI deepfake audio and video are making it hard to believe anything. It’s important to listen to the voices in your head, just like Ferrari executives did, because the company likely avoided a costly and embarrassing mistake by outwitting the fraudsters in a cynical way.
Akio Toyoda can’t blame AI for his company’s loss of credibility. No, this is a classic example of a company trying to do too much too quickly. Japanese car companies as a whole find themselves in a weird spot, which is why there are rumors of Mitsubishi joining the Honda/Nissan software alliance.
Finally, we’d like to close out this episode of The Morning Dump with a bit of news about the upcoming Cadillac Celestique.
How books can beat AI deepfakes
The scene painted by Bloomberg’s article is extremely disturbing: An unnamed Ferrari executive has been bombarded with WhatsApp messages from CEO Benedetto Vigna informing him that a big deal is coming and must be kept strictly secret.
Reportedly, this needs to be kept secret for now as it’s a bit risky and involves a lot of money. Vigna doesn’t want anyone to know about it, so he’s using a different number. Executives expect an NDA will need to be signed as soon as possible.
Of course, no one is going to want executives sending large sums of money via email and WhatsApp messages in Logano’s 2024 year, but then the phone call came from Vigna.
His imitation of Vigna’s voice was convincing, and he imitated the Southern Italian accent perfectly.
The Vigna deepfaker began explaining that he was calling from a different mobile number because he needed to discuss sensitive matters, namely a trade that may face China-related roadblocks and requires him to execute unspecified currency hedging transactions.
The executive was shocked and began to become suspicious, according to people familiar with the matter. He began to notice even the slightest robotic intonation, and his suspicions grew even more intense.
“Sorry, Mr. Benedetto, we need to verify your identity,” the executive said. He asked: What was the title of the book that Vigna recommended to him a few days ago?
To be honest, this simple article evoked a number of thoughts in my brain.
First of all, Bloomberg’s reporting is great, and secondly, there are a ton of apps out there that will create a voice AI bot in about 11 minutes, and a ton of audio released by CEOs to make this work.
Third, the world is an incredibly complicated place, full of digital malicious actors. In recent weeks, we’ve seen CDK Global hand over hundreds of millions of dollars to hackers. Five years ago, even trying to get a job at a parts store was risky.
Fourth, great job by this executive for keeping his cool fighting all of this.
Fifth and last, the irony of how all of this was resolved is almost too perfect: not only was an AI social engineering/deepfake attack repelled by a book, but the book itself is so blatant that you wonder if they were making it up: The Ten Commandments of Complexity: Acting, Learning, and Adapting in a World of Constant Change.
Here is the (translated) explanation:
Why does complexity always increase? What is the complexity dilemma? What does it mean to dance with complex systems? What is the meaning of the law of inevitable variety? What is the relationship between paradox and metamorphosis? These are some of the questions that this Decalogue seeks to answer. Within the vast amount of existing literature on complexity, the ten topics dealt with synthetically will be an opportunity for the novice reader to discover fascinating themes and for the experienced reader to reread important themes of the subject from a different perspective.
This book seems to be specifically about the complexity of business related systems, but here you have an example where complexity (in terms of communication and business structure) almost led to a significant loss of capital. Amazing!
Akio Toyoda: “If I continue to lose support, I will not be able to serve as a director next year”
Former CEO and current Chairman Toyoda Akio endured attempts to remove him from his position as director, citing all sorts of issues surrounding Toyota’s certification problems and what the company acknowledges is a strained work culture.
Toyoda himself has said he may step down next year if he is unable to turn the situation around, according to Reuters.
The 68-year-old grandson of Toyota’s founder said in an interview with the company’s own news outlet that last month’s survey results showed the lowest approval rating for directors in the company’s history.
“If things continue like this, I won’t be able to become a director next year,” Toyoda said.
Support from overseas institutional investors was particularly low, at 34%. Prior to the meeting, proxy advisory firms Institutional Shareholder Services (ISS) and Glass Lewis had raised concerns about Toyota’s handling of the certification test violations.
Given how this has played out, I think it’s premature to conclude that Toyoda is screwed — it’s hard to imagine investors would abandon him after making so much money for his family — but it’s a sign that Toyoda is at least pretending to take what happened seriously.
Mitsubishi, Nissan, Honda?
Japanese automakers have generally taken a more cautious approach to electrification, autonomous driving and software, which has led to greater success, to varying degrees, right now.
This works well for now, but what about the future? The Rivian/VW deal proves that software/autonomy is very hard and very expensive for each company to build their own version.
So it’s not a huge shock to hear that Mitsubishi, which partnered with Nissan a few years ago, may join the previously announced Honda-Nissan alliance.
The news was first reported by the Nikkei Shimbun.
Mitsubishi will work with Honda and Nissan to finalize the specifics of the partnership. The three companies aim to standardize the in-vehicle software that controls vehicles. Nissan and Honda are expected to jointly develop basic software and discuss installing it in Mitsubishi vehicles.
It would be sheer madness for Mitsubishi to try to do this on their own.
Every Celestique you see will be special.
Given all the problems GM’s current-generation Ultium EV platform caused at launch, we figured we wouldn’t see a luxury $340,000 Cadillac Celestic anytime soon.
Our friend Mark Phelan of the Detroit Free Press visited Cadillac’s new Cadillac House in Vanderbilt, Michigan, to see where fashionable people design fashionable cars to their own fashionable standards.
Every Celestiq owner meets with GM designers to choose the interior materials and look of their car, and Cadillac House is an open-air space where a variety of fabrics, leathers and woods are hidden behind wooden paneling. Cadillac House is a former cafeteria that has been renovated and expanded to include areas for customer consultations, tours and design work.
While many luxury automakers offer this service, it’s not as common among American luxury brands. We’ll also tell you a little bit about how the center got its name.
The building is named after Suzanne Vanderbilt, whom GM hired from Pratt Institute in 1955 to become one of the auto industry’s first full-fledged designers. She was part of a group they dubbed “maidens of design” for publicity purposes. Have you ever wondered how long it takes for an idea to evolve from innovative to flattering? Now you know. Though Vanderbilt was initially pigeonholed as someone who would give cars a “feminine touch,” she stayed with GM through the 1970s, working on several concept and production cars, including for Cadillac.
What I’m listening to while writing TMD
Childish Gambino, aka Donald Glover, aka Bando Stone, has a new song out, so enjoy “Lithonia” and all the things it foreshadows. Especially the horror movie twist at the end. Also, as an aside, I learned that “This is America” was originally a Drake diss song, which made me love it even more (TH from Toronto and I from Houston have both been hard on Drake).
The big question
What’s the one question someone would ask you that would indicate if you’re an AI deepfake?
