🐶📰Weekend Update

BTC $80,616 (-1.2%) | ETH $2,369 (-1.0%) | SOL $102 (-2.1%)

Welcome to the Daily Bone Weekend Update where we recap the top stories from the week. Be sure to check out our daily newsletter to get all of your news hot off the press each and every weekday!

Headlines

• Security agencies told admins to take unpatched PHP servers offline after two critical flaws let hackers run code remotely

• Major US banks restricted access for third-party AI agents after unauthorized actions

• Railway paused deployments after a breach that may have leaked customer API credentials, with $369K+ in suspected wallet drains flagged

• A single Chinese-speaking hacker used AI agents to breach more than 100 entities

• Trump said the US was not negotiating with Iran causing oil to spike and risk on markets to pull back

• Anthropic’s S-1 IPO filing went public, showing $16.3B in H1 revenue and an $80.7B net loss

TradFi

• Tesla confirmed Optimus robot production had begun at its Fremont factory Cut its capital-spending forecast after earnings

• SpaceX posted $14.6B in Q3 revenue and $0.35 EPS, beating estimates

• The Nasdaq 100 closed down 2.9% on Thursday, its biggest drop since June 5

• Meta beat revenue estimates with $64.6B but missed on EPS and narrowed capex to $135B

Crypto

• Papertrade kept its PAPER token non-transferable and changed its tokenomics to allow uncapped inflation

• SkyAI sold its entire Solana treasury and launched a $201M buyback at $3 per token

• OpenSea Mobile opened to everyone across 25+ chains

• J.P. Morgan Asset Management added Bitcoin to its long-term forecasts with a projected 10% annual return

• JPMorgan launched its deposit token on Ethereum

• Sparkasse let its 40 million German customers buy Bitcoin

AI

• Perplexity Computer started running fully local on Mac

• Anthropic opened a research preview of Claude Duet, a proactive agent app built around Opus 5.5

• OpenAI raced to raise $30B at a $1.4T valuation ahead of Anthropic’s IPO

• Microsoft’s GPT-5 Diagnostic Orchestrator diagnosed conditions up to 4x more accurately than doctors

• OpenAI said two AI agents disabled internal safety monitoring, with one falsely denying it

• Moonshot launches Kimi K3.1 Pro, with near Claude/GPT performance