Why Tesla’s new NSFW Grok AI mode has no place in a car

'Not safe for work' or 'NSFW' is a term I have come across before, but never when testing a car. Having infiltrated our mobile phones, smart devices and - to the dismay of artists - film production, AI has now made its way into our vehicles, offering some truly strange things. Volkswagen, Skoda, Mercedes-Benz and Peugeot are just a few of the manufacturers that have recently implemented ChatGPT AI, with others, including Ford, BMW and Toyota, due to join them. However, my most recent experience with AI was the most bizarre yet, as the latest Tesla Model 3 hosts the controversial Grok chatbot built by Elon Musk's xAI firm. I will start by giving the system credit where it is due: it is effective and feature-filled. It can assist with navigation, play trivia games, highlight points of interest and even read stories to your children. Offering several different voice modes, it responds almost exactly as a real person would, complete with an uncanny cadence. Other features include a language tutor and a medical support system. I even asked it to look up my own name, to which it responded with frightening detail about my career path and job description. The NSFW mode, however, is something else entirely. It's an 18+ function with several sub-settings, including ‘unhinged’, ‘motivation’, ‘argumentative’ and ‘romantic’. Reflecting the Tesla EV line-up (Model S, Model X, Model 3 and Model Y), there's also a ‘sexy’ mode. To answer one of your lingering questions: yes, the system will read you a dodgy story about a rough-and-tumble encounter between a certain big tech founder and the orange man at the top. The phrase ‘Cybertruck abs’ will never leave my mind. Thankfully, there's option for no image generation, meaning Grok's most shocking behaviour - as seen in recent news reports regarding AI-generated deepfakes and government investigations - can be avoided. The real question is: why do we need this? Does this sort of technology have a place in cars, where distractions are rampant and safety should be heavily prioritised? Absolutely not. While the public's appetite for new tech is at an all-time high, I can't help but think the blurring of lines between a safety-critical function and a boundary-pushing chatbot is a recipe for disaster. Ultimately, if this is the future of the ‘smart car’, I might just be happy staying behind the times with my aux cable and wired headphones.

main menu

Why Tesla’s new NSFW Grok AI mode has no place in a car

Musk’s chatbot is now in Model 3, offering everything from 'sexy mode' to bizarre fan fiction about tech founders

More from Autocar

Latest car reviews