However, the technology has raised privacy concerns. Jay Stanley, a senior policy analyst at the American Civil Liberties Union, warns that such technology could be used for government surveillance, corporate tracking, or stalking, as it could expose information about individuals that they never intended to share. Despite these concerns, the Stanford students believe the technology has potential beneficial applications, such as helping to identify the locations of old snapshots, conducting rapid surveys for invasive plant species, or identifying roads or power lines that need fixing.
Key takeaways:
- A student project at Stanford has developed an AI algorithm, known as PIGEON, that can accurately geolocate where photos are taken, even if it has never seen the photo before.
- The PIGEON algorithm was trained using images from Google Street View and can identify the location of a Google Street view image anywhere on earth with a high degree of accuracy.
- While the technology has potential beneficial applications, such as identifying locations of old snapshots or helping field biologists conduct surveys, there are also privacy concerns. It could potentially expose information about individuals that they never intended to share and could be used for government surveillance, corporate tracking or stalking.
- The Stanford students have held back from making their full model publicly available due to these privacy concerns. However, similar technology is likely to become widely available in the future, raising questions about how to protect privacy in the age of AI.