Click any tag below to further narrow down your results
Links
Cryptographers from Stanford and a16z address whether recent AI breakthroughs threaten the math underpinning encryption. The short answer: recent AI progress on general math problems tells us almost nothing about cryptographic security, and even smarter AI can't solve problems that don't have efficient solutions.
- Recent AI math breakthroughs didn't include any cryptography results, so they reveal nothing about whether encryption methods are actually vulnerable.
- Modern cryptography assumes certain math problems have no efficient algorithm at all—not just that humans haven't found one yet. A smarter AI can't discover an algorithm that doesn't exist.
- AI could compress the years-long vetting process for new cryptographic systems by rapidly testing whether they can be broken, building confidence faster than traditional analysis.
- Cryptographic agility—the ability to switch between different encryption methods—already exists in systems like TLS and offers defense against both AI and quantum threats.
Rhizomatica's open-source HERMES system lets remote communities send files, photos, and encrypted messages over shortwave radio instead of just voice, achieving 400-600 kilometer links with minimal infrastructure. The system proved itself when Bangladeshi fishers used it to coordinate a rescue after a boat broke down in the Bay of Bengal.
- HERMES transmits files at 400-600 bits per second over HF radio using the ionosphere as a reflector, enabling data transfer in places where terrestrial infrastructure doesn't exist
- The system handles practical emergency and logistical needs—GPS coordinates, SOS signals, medical data, supply lists, and evidence photos—faster and more reliably than voice transmission
- Bloom built optional encryption into HERMES because some partner communities in sensitive areas need protection from having their communications intercepted and weaponized
This is the App Store page for Anonine, an iOS VPN app that encrypts traffic and lets users appear to be in different countries. It offers free servers plus paid subscriptions ranging from $9.99/month to $35.99/year, though Apple notes the developer hasn't disclosed privacy practices.
- Free VPN access available with no account creation required, plus three paid subscription tiers ($9.99/month, $20.99/quarter, $35.99/year)
- Developer (Edelino Commerce Inc.) has not provided privacy policy details to Apple and will be required to do so at next app update
- Auto-renewing subscriptions with no free trial; refunds unavailable after renewal period begins
ChatGPT now lets U.S. users link Apple Health and supported medical records so the AI can compare lab results, track trends, and prep you for appointments with personalized insights. It runs on the new GPT-5.6 Sol model, adds health context to any chat with your permission, and wraps all data in extra encryption without using it to train models or target ads.
- ChatGPT can now link Apple Health and medical records (hospital systems, One Medical, Function Health) to compare lab results, flag trends, and personalize advice with user permission
- Runs on GPT-5.6 Sol for paid users and GPT-5.5 Instant for free users, with physician-run HealthBench Professional tests showing every GPT-5.6 variant beating GPT-5.5 on hard health tasks
- Health data gets extra encryption, is never used to train models or target ads, and synced data deletes within 30 days of disconnection