Discord’s May 8 Outage Was an API Error Incident—But the User Count Isn’t Official

What broke for users?

The incident affected several core parts of the Discord experience:

• Session starts and sign-ins: Discord explicitly said many users were unable to start sessions ^[15].
• Messages and app loading: Users reported failed message sends and loading problems during the outage ^[4].
• Connectivity: Reports described connection issues across regions, with one account noting concentrations in major U.S. cities including New York City, Los Angeles, San Francisco, and Seattle ^[23].
• Voice or server access: GV Wire reported user issues involving voice calls during the same outage window ^[25].

A third-party API status tracker later listed Discord as operational and said all critical functionality had recovered for users ^[22].

How many users were affected?

Discord did not publish an official affected-user count in the provided sources. The numbers circulating during the outage were public outage-report counts, not a verified total of every account, session, or device affected.

The reported figures varied by outlet and timing:

• Hindustan Times reported that Downdetector reports surged to nearly 38,000 shortly before 4 p.m. ET ^[23].
• Times Now reported nearly 60,000 outage reports just before 4 p.m. ET ^[27].
• Another Times Now report cited nearly 71,000 users or reports during the outage ^[5].
• GV Wire reported more than 95,000 users had reported problems as of 1:19 p.m. PT, citing Downdetector, which it described as tracking outages by collecting status reports from multiple sources ^[25].

The safest conclusion is that tens of thousands of people reported problems, and the highest figure in the provided source set is more than 95,000 reports ^[25]. That should not be treated as Discord’s confirmed total impact.

Timeline of the May 8 incident

Discord’s status page and third-party reporting point to a Friday afternoon disruption in U.S. time zones.

• Around 2:45 p.m. ET: Third-party reports said complaints began to spike around this time ^[23][27].
• 12:08 p.m. PDT / 3:08 p.m. ET: Discord said it was investigating errors in its API systems ^[15].
• 12:24 p.m. PDT / 3:24 p.m. ET: Discord said it had identified the issue and that many users were unable to start sessions ^[15].
• Shortly before 4 p.m. ET: Public outage reports were already in the tens of thousands, with different reports citing nearly 38,000 and nearly 60,000 reports ^[23][27].
• By 1:19 p.m. PT / 4:19 p.m. ET: GV Wire reported more than 95,000 problem reports ^[25].
• After recovery: A third-party status tracker listed Discord as operational and said critical functionality had recovered ^[22].

What the report counts mean

Outage trackers are useful for spotting widespread problems quickly, but their numbers are not the same as a platform’s internal incident metrics. A Downdetector-style count reflects submitted or collected problem reports, while Discord would be the source for a confirmed count of affected users, accounts, sessions, or infrastructure components ^[25].

So if you saw headlines saying “71,000 users” or “95,000 users,” the more precise reading is: those were outage-report figures captured at specific moments, not an official Discord user-impact total ^[5][25].

Bottom line

Discord’s May 8 outage was publicly tied to increased API errors, and those errors disrupted sessions, logins, messaging, loading, connectivity, and voice-related access for users across multiple regions ^{[4][15][24][25]}.

What remains unknown is the deeper technical root cause. Based on the provided sources, Discord identified the incident category but did not publish a detailed public explanation of what triggered the API errors, and it did not release an official affected-user count ^[3][15].