Apparently it was resolved in 2 hours.

That seems faster than usual, compare to Outlook (down for a day) and GitHub. It's still embarrassing for a heavily-relied service, but better, and I wonder if Claude (or another LLM) helped diagnose and fix the outage?

▲

isaacdl a day ago | parent | next [-]

It's back down again, new incident: https://status.claude.com/incidents/34wq7gxnxvvm

▲

armchairhacker a day ago | parent [-]

I still wonder if LLMs were used to "resolve" the first incident...

	▲	lnenad a day ago \| parent [-]
		What's there to wonder? The premiere AI company is bound to dogfood their product to help with the main problem space their product is meant to tackle.

▲

clivedup a day ago | parent | prev | next [-]

There's not many LLM provider features that operate across multiple accounts.

This allows their architecture to be simpler, and scale more easily.

GitHub meanwhile is highly relational, across repos and users, and provides real time updates over all this. Outlook is similar.

▲

cheptsov a day ago | parent | prev | next [-]

It's not resolved. For example, connectors do not work, e.g. Google Docs, etc.

▲

nxobject a day ago | parent | prev [-]

I look forward to the PR video “How we used a swarm of agents to fix an outage issue”, right after “How we used a swarm of agents to identify thousands of potential drugs for Alzheimer’s”.

	▲	armchairhacker a day ago \| parent [-]
		Judging by recent AI exploits and sandbox escapes: “how we used a swarm of agents to fix someone else’s outage”.