A carrot farm. A chocolate factory. An egg distributor. What do all these businesses have in common?
They’re all clients that the security solutions supplier eSentire has taken on in recent years.
“Everybody needs security today,” said John Moretti, a principal solutions architect for eSentire, during a recent webinar. “You can’t afford to be down. You’ll lose business.”
That’s particularly true these days because of how generative artificial intelligence (a.k.a. genAI) affects cybersecurity, presenting new opportunities as well as challenges. During the May 22 webinar co-hosted by eSentire and Stratosphere Networks, attendees learned about how generative AI is impacting the future of IT security.
For those who couldn’t attend the live event, here’s a recap of some vital information covered during the webinar. Please register here to view the recording if you want the full learning experience.
Security risks presented by generative AI
While artificial intelligence has existed since the 1950s, it’s undergone dramatic changes in the past couple of years with the explosion of new generative AI solutions. These applications produce creative content such as text, images, audio or video after absorbing patterns from input datasets.
Most of us currently use AI in one form or another daily, whether we’re reading an AI-generated summary in Google search results or asking a virtual assistant for help finding a vacation destination.
“It tries to understand and give us the best experience possible,” Moretti said of AI.
However, law-abiding citizens aren’t the only ones with the power of AI at their disposal; hackers have access to these tools, too, and they’re seizing the opportunities presented by genAI to supercharge phishing campaigns and ransomware.
“Threat actors are using AI against us every day,” Moretti said. “Threats are changing every single day across the board.”
For example, he noted that bad actors can now churn out nearly pristine phishing emails. Gone are the days when clunky grammar and abundant typos made malicious messages easy to identify.
“We actually saw a phishing attack where the email was almost totally legitimate,” he said. “There was one character that was off.”
If you’re unsure if an email is genuine, he recommended calling or texting the person and asking if they reached out: “We have to do that today. Unfortunately, there’s no way around it.”
Moretti confirmed that phishing is still the No. 1 type of cyberattack leading to ransomware incidents today, as observed by eSentire. With “ransomware as a service” packages for sale on the dark web, it’s easier than ever for malicious individuals to orchestrate attacks these days – no coding experience needed.
Additionally, eSentire has observed hackers infiltrating a business environment and biding their time for six months to a year while gathering information about the organization to construct a customized ransomware package.
In addition to helping hackers compose phishing messages and craft malware code, genAI can lead to data breaches if your employees feed it sensitive or proprietary data. For example, ChatGPT creator OpenAI acknowledge that threat actors exploited a vulnerability to access users’ chat history in May 2023, according to eSentire.
Any usage of generative AI must also comply with data privacy and security regulations such as HIPAA. Generative AI could affect your cyber insurance coverage, and eSentire notes that inputting data subject to compliance requirements into an internally hosted large language model (LLM) will subject that system to regulatory scrutiny.
How to combat evolving threats with generative AI
So, how can businesses safely deploy generative AI while addressing the risks presented by the technology? Cybersecurity providers like eSentire have developed solutions that allow you to deploy genAI applications and LLMs safely.
eSentire’s strategy involves the following:
- An AI Investigator solution that allows clients to query a proprietary eSentire LLM to learn what’s happening in their environment (e.g., how many tickets are currently open and which threats eSentire is containing)
- An LLM Gateway tool that logs data pertaining to AI usage in your environment (e.g., are your end users utilizing ChatGPT?)
- Managed detection and response (MDR) for LLM, which will be able to collect LLM interactions via eSentire’s extended detection and response (XDR) platform to support comprehensive threat detection, investigation and response. An MDR for AI solution is also in the works.
On top of that, Moretti explained that eSentire utilizes AI to search for anomalies in data gathered via their XDR platform. If the AI flags anything, the human experts in the provider’s security operations center will then make a determination and respond.
Ultimately, Moretti cautioned that you can’t completely eliminate cybersecurity risks. You can only work to close gaps, minimize exposure, and prepare for the worst so you can contain the threat and bounce back fast after a data breach.
“It’s not a matter of if,” he said. “It’s a matter of when.”
Zach Hester, a senior vice president of advisory services with Stratosphere Networks, also presented during the webinar, explaining how Stratosphere saves businesses dozens of hours by expertly identifying the best options in a vast sea of solutions. With a 100 percent objective approach, the firm’s advisors utilize extensive IT experience, partnerships with best-in-class suppliers, comparison matrices, and the latest market data to rapidly select products and services that align with each client’s specific needs.
To explore best-in-class generative AI and security solutions, connect with our advisors today by calling 877-599-3999 or emailing sales@stratospherenetworks.com.
