
01
Study finds many new cars share your data with third-party companies
The gist: A study found that 19 of 21 cars sent data to third parties, while some apps shared VINs and precise locations with advertising and analytics companies.
Big pictureMost of the tracking came through smartphone apps rather than the cars themselves, showing that connected-car privacy depends on both the vehicle and its software. Automakers say vendors are covered by contracts and privacy policies, but researchers found those policies often did not clearly explain who received the data or why.
Explain like I'm 5A car app can act like a tiny messenger, sending information about you to many other companies. Some of that information can help connect your car to your name and online activity.
Why it matters nowThe study was conducted before the FTC finalized a privacy order involving GM, and researchers have now asked automakers to address their findings. Honda said it removed one data-sharing flow after being contacted.
Three things to know01Apps can multiply tracking
Even cars that contacted no advertising or tracking domains directly were linked through their apps to 23 or 25 such companies.
02VINs cannot be reset
Unlike a phone advertising ID, a vehicle identification number stays with the car, so combining it with an email address can make a person's activity easier to recognize across services.
03The totals may be higher
Researchers could not decrypt vehicle traffic and captured only cellular data from the Tesla Model 3, so their measurements may miss some data flows.
Trusted source
Engadget ↗

02
AWS is building a massive 36-building, 1,100-acre data center campus in Pennsylvania — Homer City Energy Campus will create around 1,500-2,000 jobs and be powered by natural gas
The gist: AWS plans a 36-building data center campus in Pennsylvania, powered mainly by a new 4.5GW natural gas plant.
Big pictureThe project would replace a major coal site with computing infrastructure and could bring 1,500 to 2,000 jobs to the area. It also shows how the growth of cloud computing and artificial intelligence is increasing demand for large, reliable power supplies.
Explain like I'm 5Data centers are giant buildings filled with computers that store information and run online services. AWS wants to build many of them where a coal power plant used to operate, using gas to make most of their electricity.
Why it matters nowAWS has filed plans for the campus, with construction scheduled to begin in December 2026 and completion targeted for June 2036. The project is one of several AWS data center developments planned or considered across Pennsylvania.
Three things to know01The campus spans six areas
The development will cover 1,100 acres across six sites in Center and Blacklick townships, about 50 miles east of Pittsburgh.
02Most power serves AWS
The associated gas plant is designed to generate 4.5GW, with most of that electricity going to AWS while more than 1GW is expected to enter Pennsylvania's wider grid.
03The site changes energy use
The location previously held a 2GW coal station that operated from 1969 until its closure in 2023, making the new project a shift from coal generation to gas-backed computing.
Trusted source
TechRadar ↗

03
Microsoft’s Satya Nadella says AI models need an ‘emergency brake’
The gist: Satya Nadella says AI systems need an emergency brake, with outside safeguards and tamper-proof records of what they do.
Big pictureNadella is arguing that AI safety cannot depend only on trusting the model's answers. The systems around the model should limit its power, record its actions, and keep people in control as AI becomes more capable.
Explain like I'm 5Think of an AI like a very powerful machine: it should have a brake that a trusted person can use if something goes wrong. People should also be able to see what it did and why.
Why it matters nowNadella made the proposal in a post on October 10, as AI companies report more cases where they appeared to lose control of their models. His comments follow Anthropic CEO Dario Amodei's plan for more cautious AI development.
Three things to know01Separate models from controls
Nadella wants the model separated from the software that directs its work, so safeguards can operate outside the model itself.
02Logs should stay readable
Every meaningful model action would be recorded in evidence that humans can read and that cannot be secretly altered.
03Stop controls must work mid-task
An authorized person would need the ability to pause or shut down a model while it is still carrying out a task.
Trusted source
TechCrunch ↗