Skip to main content

The Next Cyberattack Won’t Wait. Are You Ready?

November 18, 2026

One day into the November Microsoft outage, Cofense Intelligence saw phishing emails posing as service notifications. The outage was real. The emails were designed to steal sign-in credentials. We published our findings on November 16. On November 17, CISA urged organizations to patch a critical Oracle Java vulnerability. Security teams should act on the patch guidance and keep email defenses running while they address it.

Read our November 16 Flash Alert · Read CISA’s November 17 alert

One campaign can look like many

The FBI has warned that criminals use generative AI to make scams more convincing and easier to scale. In polymorphic phishing, different versions of the same campaign can land in different inboxes. AI can help attackers produce more of those variants. Even modern filters can miss a message that’s been designed to evade AI algorithms.

For a security team, the work starts when the first report comes in. Removing that one message protects the person who reported it. Finding the other versions can protect employees who received a similar email and haven’t flagged it. That’s why teams need a way to investigate across inboxes.

Why AI alone isn’t enough

AI is essential to modern security. It processes volume and speed that no team could match by hand. But AI struggles with threats that don’t match previous patterns, and new lures are exactly what attackers are producing. Human analysts close that gap. They judge intent and context, and they respond to what automation hasn’t learned yet. Reports from employees and customers feed back into protection, so one organization’s report can help protect many others.

In one recent case, a brand-new lure got past several automated systems, including AI-based ones. Employees at a Cofense Phishing Defense Center customer reported the lure, in turn giving Cofense analysts the opportunity to protect all of our customers from this threat. Speed and human judgment working together made the difference. Cofense has seen a 22% growth from 2025 to 2026 so far in non-Microsoft related branding being used in phishing attacks as AI allows Threat Actors to expand upon the types of threats they are able to create. Depending on the industry, between 39% and 49% of phishing subject lines included the recipient’s name or other personal details. It is becoming increasingly easy for threat actors to use AI to harvest employee information and use it in phishing campaigns.

Find the campaign and remove it

Say one employee reports a fake service alert. Start with that email, then check whether other inboxes received related versions. AI helps group similar messages for analyst review. An analyst confirms which emails are malicious so the team can remove them, including copies nobody reported. That’s the difference between closing one report and removing the campaign.

See how Cofense Vision AI connects related messages

 

What organizations can do now

Walk through the steps below with the people who own email security, IT, and incident response. Start by checking who can remove a confirmed phish from multiple inboxes and how your team will coordinate if email is down.

  1. Have Cyber Security solutions that are able to be flexible in the everchanging landscape of cybercrime and AI. You should have human resources working in tandem with supervised AI to ensure that you aren't left waiting for AI security models to catch up.
  1. Map out alternative communication strategies for when you lose access to email or messaging platforms.
  2. Have and verify backups for all important company data. 
  3. Ensure your internal Phishing Training programs are properly training your employees based on real life threats and targets weaknesses in your user base.
  4. Document clear lines of communication. Employees should know who to contact when verifying a request or transaction. They should also know which backup channels to use if they suspect the original source has been compromised.
  5. Don’t rely on a single layer of defense. No filter catches everything and attackers design campaigns to bypass them. Pair your automated security tools with detection and response to ensure that someone is finding what is slipping through.
  6. Treat legitimate remote access tools with the utmost scrutiny. Attackers often abuse legitimate remote monitoring and remote management (RMM) tools, such as ConnectWise/ScreenConnect, because those tools blend in with normal IT activity. Block any RMM or remote access software your organization doesn't officially use and closely monitor the ones you do. 
  7. Patch with vetted updates. Apply tested patches for every known vulnerability. Prioritize anything actively exploited or facing the internet, like VPNs, firewalls, and remote access appliances.
  8. Shrink your attack surface. Review and remove old or unused accounts, services, appliances, and applications. Anything you don't need is just one more way in.
  9. Flag external email clearly. Use clear banners to warn users when a message comes from outside the organization. That simple visual cue helps users spot business email compromise (BEC) and impersonation attempts.

 

How Cofense helps

Cofense combines AI with human expertise to find and remove phishing from inboxes. Our Managed Phishing Defense team can help when your staff is stretched. If you’re a customer and want help reviewing your setup, contact Cofense Support.

Read The Case for Human-Supervised AI · Request a demo · Contact Cofense · support@cofense.com