Archive for November 20, 2026

Wagepoint strengthens executive leadership for its next stage of growth

Posted in Commentary with tags on November 20, 2026 by itnerd

Wagepoint has announced four executive leadership appointments, recognizing the contributions they’ve made to the business and positioning the company for growth. The appointments, effective November 1, 2026, include Jim McCreath as Chief Financial Officer, Kevin Forbes as Chief Technology Officer, Taryn Shulman as Chief Marketing Officer, and Nessa Plotkin as Vice President, Customer Operations.

Jim McCreath steps into the CFO role after six years at Wagepoint. He has played an important role in the company’s financial management and strategic decision-making, bringing financial discipline and sound judgment to key business priorities as Wagepoint has grown.

Kevin Forbes becomes CTO after more than two years leading Wagepoint’s technology organization. During that time, he has built a strong engineering team and introduced agentic engineering, strengthening how the company develops and delivers its products today and into the future. He likewise played a key role in the transition to Wagepoint’s new platform.

Taryn Shulman becomes Chief Marketing Officer. In her two years at Wagepoint, Taryn has transformed the marketing function, building a predictable performance marketing engine, leading a corporate rebrand, and sharpening market positioning. She has led go-to-market strategy for major product launches, grown Wagepoint’s partner marketing programs, and strengthened its connection with small businesses and the accounting community.

Nessa Plotkin becomes Vice President, Customer Operations. In her nearly two years, she has strengthened the support infrastructure with dedicated payroll, partner and product teams, introduced premium support, and led through the migration to Wagepoint’s new platform and two year-end seasons, keeping customers at the centre of its work.

These four appointments reflect Wagepoint’s investment in developing its leadership from within as the company continues to evolve its platform, grow its customer base, and strengthen its position in the Canadian small business payroll market.

The EU has been pwned via AI

Posted in Commentary with tags on November 20, 2026 by itnerd

The EU has been pwned by a major cyberattack. Here’s the lowdown:

The European Commission has suffered a major cyberattack, but it is not yet clear who is behind it or whether sensitive staff data was stolen.

For several days, Commission officials could not access Sysper, the internal HR platform where they check salary information, request holidays and sick leave, submit timesheets and sign up for training.

“The investigation is still ongoing. We are taking any necessary steps based on the findings as they come,” Thomas Regnier, the Commission’s spokesperson for digital, told reporters on Thursday. “We have not detected any evidence of a data breach affecting Commission staff as a whole.”

Regnier said that the attack was based on a publicly available AI tool; it was first detected on 22 October and confirmed on 22 October, and the Commission did not receive any ransomware request.

The name of the AI tool is ARTEX, which allows the identification of software vulnerabilities that can then be patched, or exploited in the case of malicious actors.

However, he declined to confirm whether it was still ongoing, nor to attribute it to a specific actor. The Commission preventively took down several of its internal apps to avoid the attack from spreading, most of which have been restored

Apparently everything is back up. But back up isn’t the point. The point is that they have been pwned. Martin Jartelius, Product Director of AI at Outpost24 had this to say:

“What’s significant about the European Commission attack isn’t that AI has invented new hacking techniques. It’s that AI can now bring existing tools, techniques and workflows together, making sophisticated attacks accessible to people who previously lacked the expertise to carry them out,”

“Open-source tools accelerate that shift because anyone can access and adapt them. But there’s an irony here: more than 1,600 instances of these hacking tools have reportedly been left exposed online, potentially turning the attackers themselves into targets. AI may be lowering the barrier to cyberattacks, but it certainly isn’t eliminating human error.”

This will sound familiar, but AI has to be part of the strategy for you to defend your organization. Because it is becoming clearer that this is really needed.

Time to patch a bunch of Nginx and PHP vulnerabilites

Posted in Commentary with tags , on November 20, 2026 by itnerd

Get ready to patch all the things again. For starters, The PHP dev team released security updates for 18 vulnerabilities. And two critical vulnerabilities in NGINX could let remote attackers run code on vulnerable web servers without logging in.

Martin Jartelius, Product Director of AI at Outpost24 had this to say:

“One serious flaw tends to bring others with it, as researchers check similar code for the same weakness. Open-source web servers and PHP are likely to see more of these issues now that AI is helping researchers find problems that have sat unnoticed in the code for years.”

So yeah. Patch all the things. But at the same time expect more of these vulnerabilities to come.

Germany raises cybersecurity threat level to critical

Posted in Commentary with tags on November 20, 2026 by itnerd

Germany’s Federal Office for Information Security (BSI) has warned (translation) that the country’s cybersecurity threat level has deteriorated from “tense” to “critical,” citing the rapid advancement of AI-powered cyberattacks and growing threats to critical infrastructure.

In its November 19 annual report, the agency said its new Cyber Index rose 50% compared with the previous reporting period.

The BSI reported an average of 160 newly disclosed vulnerabilities per day, a 35% increase year over year. The average time between vulnerability disclosure and exploitation fell from 53 days to 25 days, reaching just eight days in the second quarter of 2026.

The agency also reported a 57% increase in attacker reconnaissance activity and warned that increasingly autonomous AI agents could further accelerate attacks.

Seemant Sehgal, Founder & CEO, BreachLock

“Eight days between disclosure and exploitation is an average, and averages hide the vulnerabilities that get weaponized in hours because they sit on internet-facing systems with a public proof of concept. A 57% rise in reconnaissance tells me attackers are mapping which organizations are slow to patch before the next CVE even drops.

“Defenders who know their own exposed systems and credentials well enough to fix the reachable ones first will be fine, and everyone still working through 160 new vulnerabilities a day in order of severity score will fall behind.”

I’ll say it one more time. AI threats need to be part of your defense strategy. Plus rapid patching needs to be there to. Clearly this proves it.

AI-driven vulnerability discoveries and rushed software updates fueling outages

Posted in Commentary with tags on November 20, 2026 by itnerd

According to CyberScoop, federal officials and lawmakers are pointing to accelerated software patching as a possible contributor to recent technology outages affecting airlines, banking systems and major communications platforms.

At a November 18 congressional hearing, intelligence and homeland security officials said there was no evidence that the disruptions resulted from malicious cyberattacks. Instead, they cited the increasing pace of software updates and system upgrades.

Sen. Mark Warner warned that AI is enabling organizations to discover vulnerabilities at unprecedented speed, creating pressure to deploy patches immediately, sometimes across systems not designed to handle such rapid changes.

The concern is that emergency security updates, particularly those affecting critical infrastructure, may introduce operational failures when deployed without sufficient testing.

Ryan McCurdy, Vice President of Marketing, Liquibase

“AI is changing how quickly organizations can discover vulnerabilities but fixing them safely is still a complicated process. Engineering teams need to understand dependencies, test changes, and deploy fixes into systems that millions of people rely on. That work doesn’t become less important because the threat is urgent.

“I think the danger is treating speed as the only measure of success. A patch can close a security vulnerability and still create an outage if it breaks something else. We’ve seen several disruptions involving software defects, infrastructure changes, and maintenance, although we can’t attribute all of them to AI-driven patching.

“As we move toward an AI-driven software development lifecycle, enterprises need to automate the controls that make changes safe, not bypass them. For production databases, that means validating changes, enforcing policies, requiring approvals when necessary, and maintaining audit evidence of what happened. The goal shouldn’t be to patch slower. It should be to make safe changes faster.”

ㅤ

Darin Fredde, Senior Director and Offensive Security Evangelist, Ridge Security

“Accelerated patching is a plausible explanation, but it has a familiar ring to anyone who’s heard an airline announce a ‘software glitch.’ I’d give that explanation more weight once we know what changed, how the failure spread, and why the fallback didn’t hold.

“When a handful of shared providers can disrupt airlines, banks, and communications platforms, I think about the A in the CIA triad: availability. AI can help teams find flaws faster, and some flaws genuinely warrant emergency action. The discipline is knowing which fixes need to go out now, rolling them out in stages where possible, and having a tested way back. Otherwise, it starts to look a little like the dodos in Ice Age: everyone’s focused on saving the watermelon, and the cliff sneaks up on them.

“The practical question I’d ask is whether we can still deliver essential services if a shared provider goes down. Can we route around the failure? Can staff keep critical work going offline if a cloud service is unavailable? Those are things we can test today, even while the incident reviews are still underway. That’s where I’d put the next dollar. One question I’d add to any outage drill is whether you can still test your defenses with the internet connection down. That’s one reason we built RidgeOne. It runs security tests and the AI model locally, so teams can keep testing inside their own network without a live connection to a cloud service. The cloud can go down. Your security testing can keep going.”

Organizations would do well to get ahead of this by not trusting AI as they are now. Because trusting AI clearly isn’t working.