After reading @drakefjustin ’s and @VitalikButerin 's takes, here are a few key points, along with my own thoughts: ■ AI is speeding up math research, which could challenge cryptographic assumptions sooner than expected. It’s worth planning ahead, calmly. ■ ECDSA and elliptic curves aren’t the only concern. Lattice-based systems like ML-DSA, Falcon, FHE and LWE could also face new risks. ■ Where possible, hash-based cryptography like WOTS and SPHINCS may be a stronger long-term option, since hashes have less mathematical structure to exploit. ■ Public-key encryption can’t rely on hashes alone, so it still needs structured systems. That means we may need larger keys and more conservative security parameters. ■ No need to panic or rush into moving funds. If it’s easy, using fresh addresses that have never signed a transaction is a sensible precaution. ■ Be careful with migrations: a botched move can cost more than a hack. ■ For multisig, off-chain confirmations can help keep signer signatures out of public view. ■ For privacy protocols, avoid putting encrypted notes on-chain when there’s a practical off-chain alternative. My take: the truth is probably somewhere in the middle. The risks deserve serious attention, but today’s warnings are still forecasts, not proof that these systems have been broken.
The current AI narrative is shifting toward Personal AI Agents. Not just chatbots, but agents that can remember, act, use tools, run tasks, and work across your digital life. I compared 4 different approaches 👇 One approach stands out to me: local-first AI agents. AgentOS is built around this idea. Your agent runs locally, and your data stays under your control. It is not just a personal agent you use, but one you actually own. imo, Personal AI Agents could be the next big trend. The next wave isn’t just better models. It’s AI that actually gets things done.
Underdog by @0xSigil is an impressive step toward fast, private, on-device personal AI. It makes everyday work local with its own inference stack. AgentOS goes further as a provider-neutral runtime for agents: Pilot Router, minimum-token workflows, persistent memory, multi-channel execution, approvals, and native agentic trading. Different layers, complementary strengths. See the diagram below for the full comparison 👇 #AI #Agents
More money will keep getting printed, and it will keep losing value. Meanwhile, good assets will keep going up. That is why we invest. If you pick the right assets, you profit. Do not waste your time on pure gamble plays. Focus on solid projects that help you stack wealth over time.
