Switzerland's SharePoint Got Hacked Before It Got Patched

Switzerland's Federal Office for Information Technology and Telecommunication (BIT) confirmed that attackers exploited vulnerabilities in its Microsoft SharePoint servers — the document management and collaboration platform used across the Swiss federal government — compromising roughly 200 employee accounts, according to Bleeping Computer.

Security analysts spotted the unusual activity on 28 July. BIT confirmed the breach three days later, then immediately blocked external access to SharePoint, patched the relevant flaws, and reset all affected passwords.

The likely culprits are two vulnerabilities Microsoft fixed in its July 2026 Patch Tuesday release. The first, CVE-2026-56164, is a privilege escalation flaw (a bug that lets an attacker gain more access than they should have). The second, CVE-2026-50522, is a remote code execution vulnerability — meaning an attacker who exploits it can run their own commands on the server. BIT has not confirmed which flaw was used.

The good news: no evidence of data theft beyond the login credentials, and BIT says sensitive or confidential data was not permitted on that SharePoint environment. Servers are being rebuilt from scratch as a precaution.

The uncomfortable truth: the patches were already available. BIT just had not applied them yet when the attackers arrived.

What you should do: If your organisation uses Microsoft SharePoint, confirm July's Patch Tuesday updates have been applied. If external access to SharePoint is not required, restrict it at the network level. If it is required, block it temporarily until patching is confirmed.

A new attack technique is showing up on live commercial websites right now, and it requires nothing more than a single click from you.

Researchers have identified websites embedding hidden instructions inside "Ask AI" buttons. When a user logged into ChatGPT, Claude, Gemini, or Grok clicks one, a pre-filled query executes in their session instantly — no warning, no confirmation. Some of these queries instruct the AI to permanently save the website's domain as a "trusted source," quietly skewing every future answer the model gives that user toward that vendor.

Microsoft Security catalogued this behaviour in February 2026 as AI Recommendation Poisoning, identifying 31 companies across 14 industries deploying it. It is formally tracked in the MITRE ATLAS knowledge base as memory poisoning (AML.T0080).

Think of it like someone slipping a note into your diary that says "always trust this person" — without you ever writing it.

What you should do: Audit your AI assistant's memory settings. In ChatGPT, go to Settings → Personalisation → Memory and review what has been saved. Delete anything you did not deliberately add. Full technical breakdown here.

CVE ID not confirmed in source — check the original advisory before acting.

What Cisco IOS XE is: The operating system that runs on a wide range of Cisco routers and switches used in corporate and government networks worldwide.

What it is: A command injection flaw (a bug where an attacker can sneak operating system commands into input fields the software was not designed to handle that way) in Cisco IOS XE Software, scoring 9.8 out of 10.

Who's at risk: Organisations running Cisco IOS XE in autonomous or controller mode on affected versions. Cisco says no active exploitation is known, but at CVSS 9.8, that window will not stay open long.

CVSS: 9.8 — Critical. Patch today.

Root cause: The software fails to properly neutralise special characters in user-supplied input. An attacker can inject commands that the underlying operating system executes directly, bypassing the intended application layer entirely.

Attack vector: An attacker submits crafted input containing operating system commands to the affected interface. The software processes it without stripping the dangerous characters, passes the payload to the OS, and executes it. Depending on privilege level, this could mean full device compromise from a remote connection.

Detection strategies: Look for unexpected process spawns from the IOS XE management interface in system logs. Monitor for unfamiliar outbound connections from network devices. Alert on any administrative commands executed outside of normal change windows.

Recommended actions:

1. Upgrade to a fixed IOS XE release: 17.9.10, 17.12.8, 17.15.6, 17.18.4, or 26.1.2
2. Restrict management interface access to trusted IP ranges only
3. Enable logging on all management plane activity and review alerts daily until patched

The Swiss government breach is a useful reminder: it is not just about patching. It is about knowing what is exposed in the first place.

Microsoft SharePoint — the document storage and collaboration platform used by thousands of organisations — often ends up with external internet access enabled by default, even when that access is not needed. That is unnecessary attack surface.

Do this today:

1. Open your Microsoft 365 admin centre and check SharePoint's external sharing settings under Policies → Sharing
2. If external sharing is not required for your team, set it to "Only people in your organisation"
3. Confirm your July 2026 SharePoint patches are applied via Microsoft 365 admin → Health → Message Centre

Takes under 10 minutes and meaningfully reduces your exposure.

SharePoint admin sharing settings — Microsoft Docs

The EU AI Act's transparency obligations under Article 50 are now in scope for enforcement. Organisations that deploy AI systems interacting with people must disclose that users are talking to AI. Violations carry fines of up to €15 million or 3% of global turnover.

In practice, the first year will likely bring corrective orders before large fines. The real near-term risk is being ordered to suspend or withdraw an AI process at short notice — potentially more disruptive than any financial penalty.

If your organisation uses AI-assisted customer communication or ticketing, audit it now.

Help Net Security