Aug 1, 2026 · 21 min · 12 segments
On April 15, 2026, Anthropic's Claude chatbot, Claude Code, and API went sideways for nearly three hours, with elevated error rates starting at 14:53 UTC and full resolution not coming until 17:42…
Because today we're looking at April 15th, 2026.
And I want you to just picture this for a second.
You are in the middle of your workday.
Right.
Just a normal Wednesday.
A completely normal Wednesday.
Your team is humming along.
They're generating code, maybe drafting client emails, summarizing massive data sets.
And then suddenly nothing.
Just completely dark.
Completely dark.
And, you know, I get that three hours might not sound like a lifetime.
Yeah.
In the grand scheme of things, it's a blip.
But we really need to talk about what those three hours actually represent.
We are living in a reality now where AI is no longer just like a shiny novelty or a fun party trick for your marketing team.
It is critical infrastructure now.
Absolutely.
It is.
And our mission today is to figure out exactly what happened on April 15, why AI failures are fundamentally different from traditional cloud outages, and most importantly, how you can protect your own business operations when this happens again.
Because today we're looking at April 15th, 2026.
And I want you to just picture this for a second.
You are in the middle of your workday.
Right.
Just a normal Wednesday.
A completely normal Wednesday.
Your team is humming along.
They're generating code, maybe drafting client emails, summarizing massive data sets.
And then suddenly nothing.
Just completely dark.
Completely dark.
And, you know, I get that three hours might not sound like a lifetime.
Yeah.
In the grand scheme of things, it's a blip.
But we really need to talk about what those three hours actually represent.
We are living in a reality now where AI is no longer just like a shiny novelty or a fun party trick for your marketing team.
It is critical infrastructure now.
Absolutely.
It is.
And our mission today is to figure out exactly what happened on April 15, why AI failures are fundamentally different from traditional cloud outages, and most importantly, how you can protect your own business operations when this happens again.
The rest of this transcript — segmented and speaker-labeled, so you land on the exact moment something was said
Search every transcript — by keyword, by phrase, or by meaning, across every show Radar indexes
Trends — what is surging across podcasts, measured against its own baseline
Alerts — when a name you follow appears in a newly indexed episode
No account is needed to search Radar.