Bitcoin Holds $80K as ETF Outflows, AI Security Rules and Institutional Buys Converge

Bitcoin traded near $80,800 on Saturday morning after a turbulent week that pushed the leading cryptocurrency below its 50-day moving average for the first time in 66 consecutive sessions. The slide followed a broader tech-stock selloff, with the Nasdaq posting its largest single-day decline since June as investors rotated out of growth assets. Analysts now watch $80,000 as a critical demand floor, with the 100-day moving average sitting at $75,291 below it.

Spot Bitcoin ETFs amplified the pressure with a wave of redemptions. According to The Block, Bitcoin ETF products swung back to net outflows, recording roughly $636 million in withdrawals that erased the prior session's inflows entirely. The sharp reversal in institutional fund flows rattled short-term sentiment and reinforced caution among traders eyeing resistance at the 20-day moving average near $82,560.

Institutional Wallets and Bitcoin Core Both Make Moves

Despite the ETF selling, on-chain data tracked by Investing.com showed large holders moving coins off exchanges. Fidelity's Bitcoin ETF wallet reportedly received 282.63 BTC worth roughly $23M from Binance, while BlackRock-linked wallets took in another 384.77 BTC valued at approximately $31.4M. Net spot exchange outflows hit around $465.55M on October 22 alone, a signal some analysts read as accumulation by longer-term holders at current price levels.

Separately, Bitcoin Core developers merged a significant security policy change on October 22. As CoinDesk reported, the project will now rapidly disclose and patch any vulnerability that a large language model can surface with a basic prompt, ending embargoes that previously stretched up to a year. The shift follows a volunteer Red Team exercise in late July in which 16 developers used AI models to scan 390 Bitcoin projects in roughly a day, logging 4,962 findings with 85 classified as critical. Maintainers say the old strategy of quiet patches followed by delayed disclosures no longer makes sense when AI tools can independently rediscover the same flaws before node operators have time to upgrade.