All vulnerabilities
HIGHOpSeccurated

OPSEC-TWITTER-2020

Social media · Twitter

Summary

On 15 July 2020, the Twitter accounts of Barack Obama, Joe Biden, Elon Musk, Bill Gates, Jeff Bezos, and Apple all tweeted the same thing: send Bitcoin and I will send back double. It was a scam, and it ran from inside Twitter. Attackers had phoned a handful of Twitter employees, posed as IT, and talked them out of their credentials, which gave access to an internal admin tool that could take over any account on the platform. The mastermind turned out to be a 17-year-old. It is the lesson that a powerful internal "god-mode" tool is only as secure as the most socially-engineerable employee who can reach it.

How it happened

The attack was a coordinated phone spear-phishing (vishing) campaign that began on 14 July 2020. The attackers gathered details on their targets, called them while impersonating Twitter IT, and socially engineered them into handing over their credentials, capturing the login (and the multi-factor prompt) on a fake VPN page in real time. They first phished lower-level employees who could not reach the admin panel, then used those footholds to navigate internal systems and find staff who could. Those credentials unlocked the real prize: Twitter's internal account-management tool, a "god-mode" panel that could change any account's associated email address and reset or bypass its two-factor authentication.

With that tool, the per-user security of even the most famous accounts was irrelevant. The attackers targeted 130 accounts, tweeted from 45, accessed the direct-message inboxes of 36, and downloaded the full account archive for 7. From verified accounts belonging to presidents, billionaires, and major companies, they ran a simple Bitcoin-doubling scam and pulled in about $118,000 in a few hours before Twitter scrambled to lock things down, briefly preventing all verified accounts from tweeting at all.

The damage

The financial take, about $118,000, was almost the least alarming part. The same access could have moved markets or sparked a geopolitical incident, a fake announcement from a world leader's account, and the fact that it was "only" a crypto scam was something close to luck. Four people were ultimately charged, including the 17-year-old mastermind, Graham Ivan Clark, who pleaded guilty to a three-year sentence; a fourth defendant, the UK's Joseph O'Connor, was later extradited and sentenced to five years in US prison in 2023. New York regulators also found that Twitter had gone without a chief information security officer for the seven months before the hack. The episode exposed how few people, and how little protection around a single powerful tool, stood between social engineering and total control of the platform.

Why Twitter 2020 still matters

It is the internal-admin-tool plus social-engineering lesson. A "god-mode" tool that can take over any account is an enormous concentration of power, and if a phished employee can reach it, every user's own two-factor authentication is beside the point. The defences: require phishing-resistant MFA for employees, put powerful internal tools behind least privilege and multi-person approval for high-impact actions like account takeover, log and alert on every use of that tooling, and train staff against phone impersonation. It is the same social-engineering-to-internal-tools pattern later seen at MGM and Caesars.

How to fix it

  • Cut access to the internal admin tool and reset credentials and sessions for affected employees; assume the tool was fully abused.
  • Re-secure hijacked accounts (reset emails, force re-authentication) and review what account data was downloaded.
  • Add approval, logging, and alerting to the admin tool before re-enabling it, so no single account can take over users silently.

How to avoid it

  • Require phishing-resistant MFA (FIDO2/passkeys) for all employees; the entry point was phished credentials and a relayed MFA prompt.
  • Put powerful internal tools behind least privilege, granular permissions, and multi-person approval for high-impact actions like account takeover.
  • Log and alert on every use of admin or god-mode tooling, and rate-limit sensitive actions.
  • Train staff against phone (vishing) impersonation, and give them a way to verify "IT" callers out of band.

References

Related vulnerabilities

All OpSec →