yes sorry, just updated my comment shortly before you replied.
This is CVE-2025-36911, the other ones were CVE-2025-20700, CVE-2025-20701, CVE-2025-20702. Coincidentally a similar set of headphones affected.
This one also has a pairing vulnerability, but I assume fast pair is on the BLE level:
> To start the Fast Pair procedure, a Seeker (a phone) sends a message to the Provider (an accessory) indicating that it wants to pair.
> [...] allowing unauthorised devices to start the pairing process [...]
It's a pity that this is only awarded with $15k, this is a really bad vulnerability - which clearly required thoughtful investigation, publishing, reporting, ... and would have a much bigger audience in the exploit market.
Would it be be trivial to have a init container to do CA injection? Maybe though mutating admission controller? Then some CNI magic to redirect outbound traffic to do transparent proxying?
Just because you cannot see how a vulnerability can be exploited does not mean that others can. As you describe, people seem to assume that the only way the config file ends up on the server is «physically» editing it.
An anecdote: I have been struggling with exploiting a product that relies on MongoDb, I can replace the configuration file, but gaining RCE is not supported «functionality» in the embedded version as the __exec option came in a newer version.
It is strange how EVs are measured by how far they can go full charge when this is a metric I never have seen for fossile cars. It tells a story how inconvenient EVs or the charging network really is
It also tells the story that the energy price per mile is insignificant. Vehicles that use gas advertise miles per gallon because it significantly affects costs. They also advertise range (or maybe fuel tank capacity), but not as prominently.
An electric car's miles per energy is not as relevant, because current electricity prices are sufficiently low such that people won't really care whether it's 3 miles per kWh or 5 miles per kWh. They will care about how far they can go on a single charge, hence range is a metric that is often advertised.
You are actually more likely to buy a car just after you have bought a car than the 10 years you did not need to buy a car. Maybe not cars, but I’ve heard this argument for kitchen appliances. If you for some reason return the item you just bought, you may buy what you get ads for. Maybe you regret you did not get the premium one, especially when they shove it in your face afterwards…
Appliances, sure, because having bought a new blender I might be tempted to look at replacing that old toaster as well. I'm clearly in an appliance-buying mood, and if I'm not, maybe I can be persuaded in that direction.
Cars? People who just bought a car are generally upside-down, and will not be looking to trade or buy another anytime soon.
I feel like I'm far too eager to accept whatever I bought, and reluctant to return it. Maybe I should play their game and return more stuff when it's not quite perfect.
reply