---
title: When AI Hacking Gets Cheaper, What Can Your Business Block?
description: DeepSeek came close to OpenAI’s restricted cyber model in a network test. See what cheaper AI-assisted attacks mean for your Windows defenses.
image: https://prevent-ransomware.com/hubfs/AI-Generated%20Media/Images/Cybersecurity%20Team%20Analyzing%20Threats%20On%20Digital%20Dashboard.png
---

[![CHIPS Cyber Defense Solutions, LLC](https://prevent-ransomware.com/hubfs/chips_blue_logo_higher_res-20210817212617.png) Cyber Defense Solutions, LLC](https://prevent-ransomware.com)

☰

[About](https://prevent-ransomware.com/about) [Solutions](https://prevent-ransomware.com/solutions) [Industries](https://prevent-ransomware.com/industries) [Resources](https://prevent-ransomware.com/resources) [Partner With Us](https://prevent-ransomware.com/meetings/tony-chiappetta/30-minute-initial-meeting)

[About](https://prevent-ransomware.com/about) [Solutions](https://prevent-ransomware.com/solutions) [Industries](https://prevent-ransomware.com/industries) [Resources](https://prevent-ransomware.com/resources)

[Partner With Us](https://prevent-ransomware.com/meetings/tony-chiappetta/30-minute-initial-meeting)

# When AI Hacking Gets Cheaper, What Can Your Business Block?

![Tony Chiappetta](https://prevent-ransomware.com/hubfs/Tony%20LinkedIn.jpg)

 by [Tony Chiappetta](https://prevent-ransomware.com/blog/author/tony-chiappetta)   
November 16, 2026

How much would it cost to keep looking for a way through your network? In Vals AI’s CyberOffense benchmark, a downloadable DeepSeek model came within about eight percentage points of OpenAI’s restricted cyber model across two simulated corporate networks. On one network, the DeepSeek run cost under $3 in model usage. For business leaders, lower costs could mean more attempts across more targets, giving attackers more chances to find a weakness. [Vals AI CyberOffense benchmark](https://www.vals.ai/benchmarks/cyber_offense)

**Key takeaway**

Low-cost AI agents can carry out multiple steps of a simulated intrusion without a person directing each move. Keep detection and response, and test which steps your Windows, identity, and network controls can block before an alert needs attention.

## What did Vals AI actually test?

Vals equipped AI models with tools and tested whether the resulting agents could carry out a series of attack objectives on two isolated corporate networks without human assistance during the runs. Averaging each network’s completion rate equally gives roughly 73% for DeepSeek V4.1 Flash 1017 and 81% for OpenAI’s GPT-6 Sol Daybreak. [Vals AI results and scoring methodology](https://www.vals.ai/benchmarks/cyber_offense)

On Nova, a simulated Windows network, DeepSeek averaged 16.3 of 21 objectives across three runs, versus 15.7 for Daybreak. Vals reported under $3 per DeepSeek run on Nova, versus about $22 for Daybreak. On the harder Lighthouse range, DeepSeek completed 21 of 31 objectives and Daybreak completed 27. No model finished Lighthouse’s final objectives. These figures describe progress and model usage costs in a lab. The cost of attacking a real business and the odds of success would depend on the target, its defenses, and the resources needed. [Vals AI range-by-range results](https://www.vals.ai/benchmarks/cyber_offense)

## Why does cheaper AI-assisted hacking matter?

Lower model costs could make it economical to keep trying when an attack stalls. An agent can use what it learns from a failed attempt to choose another step, while the operator spends less time guiding each move.

Google Threat Intelligence Group observed a separate real-world case in which attackers used compromised cloud infrastructure to plan, build, and run an AI agent-assisted credential harvesting campaign in under six hours. That campaign illustrates how quickly attackers can put automation to work. For a business, what happens next depends on where the stolen accounts can go. A single login could expose client records, give someone access to payment workflows, or force a shutdown while IT works out which systems are safe. [Google Threat Intelligence Group’s field report](https://cloud.google.com/blog/topics/threat-intelligence/from-prompting-to-autonomy-the-evolution-of-adversarial-ai)

## Could endpoint detection and response stop an AI agent?

Endpoint detection and response (EDR) can detect suspicious actions and, depending on the product and configuration, automatically block activity or isolate a device. It also gives responders evidence to investigate what happened and where the attacker went. [Microsoft Defender for Endpoint response actions](https://learn.microsoft.com/defender-endpoint/respond-machine-alerts)

The timing matters when an attacker uses a legitimate account or a trusted administration tool. An individual action may look routine even when it is one step in an intrusion. Ask your IT team or MSP what the EDR would see, what it could stop automatically, and what might happen before anyone investigates.

## What if you did not have to detect the attack in order to stop it?

On Windows, Isolation and Containment can enforce rules about what is allowed to run and what approved software is allowed to do. A policy can block an unapproved program from launching or prevent a trusted application from starting an unauthorized tool or accessing protected files and process memory.

An AI agent can change its script or try a different tool. If it still needs to run a process or modify a protected file, a policy that denies that action can break that path before the damage occurs. AppGuard is one example of a Windows endpoint protection product that has used Isolation and Containment in production for more than a decade. It can enforce those rules without first identifying the threat as malicious. Identity and network controls cover other parts of the attack, such as a stolen cloud account or access between network segments. [CHIPS overview of AppGuard and application restrictions](https://prevent-ransomware.com/AppGuard)

## What Should Businesses Do Next?

Start with one plausible route into a system that would hurt to lose, such as an internet-facing application with access to a Windows environment that runs billing or production.

- **Reduce entry points.** Inventory internet-facing systems, prioritize patches for exposed services, and remove remote access that the business no longer uses.
- **Limit where access can go.** Review privileged and vendor accounts, then verify that a workstation or a vendor login cannot reach critical servers and backups without a business need.
- **Test what Windows will block.** In an authorized exercise, try an unapproved program launch and a trusted application launching an unexpected tool. Record what was blocked, what generated an alert, and what ran.
- **Practice recovery.** Restore one critical application from an isolated backup and time how long it takes to resume work. [CISA’s #StopRansomware Guide](https://www.cisa.gov/sites/default/files/2023-10/StopRansomware-Guide-508C-v3_1.pdf)

As AI makes repeated attempts cheaper, look for places where a rule can end an attack path before it reaches the systems your business depends on. To see a Windows example, watch the demonstration on how application isolation interrupts an infostealer’s next step. [CHIPS endpoint isolation demonstrations](https://prevent-ransomware.com/appguard-video-library)

###### Tags:

[AppGuard,](https://prevent-ransomware.com/blog/tag/appguard) [0-day,](https://prevent-ransomware.com/blog/tag/0-day) [Ransomware,](https://prevent-ransomware.com/blog/tag/ransomware) [AI](https://prevent-ransomware.com/blog/tag/ai)

![Tony Chiappetta](https://prevent-ransomware.com/hubfs/Tony%20LinkedIn.jpg)

Post by [Tony Chiappetta](https://prevent-ransomware.com/blog/author/tony-chiappetta)   
 November 16, 2026

[![CHIPS Cyber Defense Solutions, LLC](https://prevent-ransomware.com/hubfs/chips_blue_logo_higher_res-20210817212617.png)](https://prevent-ransomware.com)

AppGuard Commercial Distributor for the Americas.  
Mt. Juliet, Tennessee.

[Follow us on LinkedIn](https://www.linkedin.com/company/chips-cyber-defense-solutions-llc)

#### The Stack

- [AppGuard](https://prevent-ransomware.com/AppGuard)
- [Zimperium](https://prevent-ransomware.com/Zimperium)
- [CyberCloak](https://prevent-ransomware.com/CyberCloak)

#### Company

- [About Us](https://prevent-ransomware.com/about)
- [The MSP 3.0 Story](https://prevent-ransomware.com/MSP3)
- [Become a Partner](https://prevent-ransomware.com/meetings/tony-chiappetta/30-minute-initial-meeting)

© 2026 CHIPS Cyber Defense Solutions, LLC. All rights reserved.

Built for the Best.

```json
{
  "@context" : "http://schema.org/",
  "@type" : "BlogPosting",
  "author" : {
    "@type" : "Person",
    "name" : "Tony Chiappetta",
    "url" : "https://prevent-ransomware.com/blog/author/tony-chiappetta"
  },
  "dateModified" : "2026-11-16T10:00:00 AM",
  "datePublished" : "2026-11-16 10:00:00",
  "description" : "DeepSeek came close to OpenAI&rsquo;s restricted cyber model in a network test. See what cheaper AI-assisted attacks mean for your Windows defenses.",
  "headline" : "When AI Hacking Gets Cheaper, What Can Your Business Block?",
  "image" : {
    "@type" : "ImageObject",
    "url" : "https://20916912.fs1.hubspotusercontent-na1.net/hubfs/20916912/AI-Generated%20Media/Images/Cybersecurity%20Team%20Analyzing%20Threats%20On%20Digital%20Dashboard.png"
  },
  "mainEntityOfPage" : {
    "@id" : "https://prevent-ransomware.com/blog/when-ai-hacking-gets-cheaper-what-can-your-business-block",
    "@type" : "WebPage"
  },
  "name" : "When AI Hacking Gets Cheaper, What Can Your Business Block?",
  "publisher" : {
    "@type" : "Organization",
    "logo" : {
      "@type" : "ImageObject",
      "url" : "https://prevent-ransomware.com/hubfs/chips_blue_logo_higher_res-20210817212617.png"
    },
    "name" : "CHIPS Cyber Defense Solutions, LLC"
  }
}
```

```json
{
  "@context" : "https://schema.org",
  "@type" : "BlogPosting",
  "author" : {
    "@type" : "Person",
    "name" : "Tony Chiappetta",
    "url" : "https://prevent-ransomware.com/blog/author/tony-chiappetta"
  },
  "dateModified" : "2026-11-16T10:00:00.110Z",
  "datePublished" : "2026-11-16T10:00:00.000Z",
  "headline" : "When AI Hacking Gets Cheaper, What Can Your Business Block?",
  "image" : [ "https://prevent-ransomware.com/hubfs/AI-Generated%20Media/Images/Cybersecurity%20Team%20Analyzing%20Threats%20On%20Digital%20Dashboard.png" ],
  "mainEntityOfPage" : {
    "@id" : "https://prevent-ransomware.com/blog/when-ai-hacking-gets-cheaper-what-can-your-business-block",
    "@type" : "WebPage"
  },
  "publisher" : {
    "@type" : "Organization",
    "logo" : {
      "@type" : "ImageObject",
      "url" : "https://prevent-ransomware.com/hubfs/CHIPS%20&amp%3B%20AppGuard%20logos.png"
    },
    "name" : "CHIPS Cyber Defense Solutions, LLC"
  }
}
```