AvalonX
AI-powered continuous penetration testing for small and midsize businesses.
Website: https://www.avalonx.ai/
Cover Block
Open sources
| Field | Value |
|---|---|
| Name | AvalonX |
| Tagline | AI-powered continuous penetration testing for small and midsize businesses [Associated Press, October 2026] |
| Headquarters | Alpharetta, Georgia, US |
| Founded | 2025 |
| Stage | Seed |
| Business Model | SaaS |
| Industry | Security |
| Technology | AI / Machine Learning |
| Geography | North America |
| Growth Profile | Venture Scale |
| Founding Team | Co-Founders (3+), Craig Goodwin, Ricardo Valdes, Ben O’Nion [Associated Press, October 2026] |
| Funding Label | Seed |
| Total Disclosed | Up to $10,000,000 [Associated Press, October 2026] |
Links
Open sources
- Website: https://www.avalonx.ai/
- Product page: https://www.avalonx.ai/
- Team page: https://www.avalonx.ai/meet-the-team
- Funding news: https://www.avalonx.ai/news-funding
What an Investor Needs First
PUBLIC AvalonX is building continuous, AI-powered penetration testing for small and midsize businesses, and it merits attention now because it surfaced from stealth in October 2026 with a disclosed funding commitment of up to $10 million from Ridown Capital and a product aimed at lowering the cost and cadence barrier of traditional pen tests [Associated Press, October 2026] [Citybiz, October 2026]. The near-term story is straightforward: the company says it spent 18 months in stealth before launch, then entered the market with Artemis, a subscription product that tests domains, applications, IP ranges, cloud accounts, and internal networks rather than relying on one-off annual assessments [Associated Press, October 2026].
The technical pitch rests on automation with a human backstop. According to launch coverage, Artemis uses autonomous agents to simulate attacker behavior, identify and validate vulnerabilities, map potential lateral movement, and generate remediation guidance, while human penetration testers review findings that need additional expert verification [Associated Press, October 2026] [Pulse 2]. AvalonX also says the system is powered by Aether, a proprietary 400-billion-parameter security AI trained on nine years of offensive-security data and supported by 60 billion dark-web records, although the model-scale claim is still primarily supported by company-linked or launch-distribution materials rather than independent technical validation [Associated Press, October 2026] [avalonx.ai] [Pulse 2].
The founding team gives the company some early credibility in a market where trust matters more than a polished demo. CEO and co-founder Craig Goodwin previously held CISO roles at Fujitsu, Monster, and CDK Global, and the company says its three co-founders, Craig Goodwin, Ricardo Valdes, and Ben O'Nion, collectively bring more than 50 years of information-security and technology experience [Associated Press, October 2026]. Commercially, AvalonX appears to be pursuing a SaaS model, with website pricing that says continuous testing plans start at $249 per month, though that entry-point figure is company-published and should be treated as directional until customer packaging and realized contract values are clearer [avalonx.ai].
What matters over the next 12 to 18 months is less the launch announcement than proof that the product can convert an accessible price point into durable usage, credible detection outcomes, and referenceable SMB adoption. Public materials support the same-day initial-results claim and the no-installation workflow for external assessments, but there is still no public evidence of named customers, deployment scale, renewal behavior, or independent benchmarking against incumbent pentest vendors such as Cobalt [Citybiz, October 2026] [Pulse 2]. On the present record, AvalonX looks like an experienced-team, early-stage bet on making offensive security more continuous and more purchasable for smaller companies, with the central diligence question being whether the automation layer drives materially better economics without sacrificing trust [Associated Press, October 2026] [Citybiz, October 2026].
Partially corroborated -- Core company, product, founder, and funding facts are supported by Associated Press and Citybiz, but pricing and some model-specific claims rely on company materials.
Taxonomy Snapshot
| Axis | Value |
|---|---|
| Stage | Seed |
| Business Model | SaaS |
| Industry / Vertical | Security |
| Technology Type | AI / Machine Learning |
| Geography | North America |
| Growth Profile | Venture Scale |
| Founding Team | Co-Founders (3+) |
| Funding | Seed, total disclosed about $10,000,000 |
Inside the Company
PUBLIC AvalonX appears to be a newly formed cybersecurity company in Alpharetta, Georgia, focused on continuous penetration testing delivered as software for small and midsize businesses [avalonx.ai, retrieved 2026]. The public record here is still thin, which matters because the company only emerged from stealth in late October 2026 and much of what is knowable today comes from its own website and launch materials [avalonx.ai, retrieved 2026]. On the facts that are visible, AvalonX was founded in 2025, positions itself as a SaaS security business, and is operating from North America with a venture-scale posture rather than a services-only consultancy model [avalonx.ai, retrieved 2026].
The timeline is short but clear enough to sketch. AvalonX says it spent 18 months in stealth before announcing its launch on October 28, 2026, and its website ties that launch to the debut of its Artemis platform and the naming of co-founders Craig Goodwin, Ricardo Valdes, and Ben O'Nion [avalonx.ai, retrieved 2026]. The same public materials place the company in Alpharetta and frame the initial commercial offer as subscription-based continuous testing, with entry pricing beginning at $249 per month according to the website [avalonx.ai, retrieved 2026].
A few points remain outside the public file. No legal entity name or state filing appears in the supplied source set, and no accelerator affiliation is confirmed in the available materials [avalonx.ai, retrieved 2026]. That leaves the company overview anchored mostly to first-party disclosures, but the chronology itself is internally consistent: founded in 2025, operated in stealth for roughly 18 months, then launched publicly in October 2026 from the Atlanta-area base in Alpharetta [avalonx.ai, retrieved 2026].
Claim stands unchecked -- This section relies primarily on company website materials, with no state filing or other independent corporate record included in the supplied sources.
Under the Hood
MIXED The product proposition is straightforward enough to matter: AvalonX says it sells continuous, AI-powered penetration testing for small and midsize businesses, a buyer set that often purchases periodic testing rather than a standing offensive-security program [Associated Press, October 2026]. At launch, the company described Artemis as a system that uses autonomous agents to simulate attacker behavior, identify entry points, validate vulnerabilities, map lateral movement, and generate remediation guidance, with human penetration testers reviewing findings that require expert verification [Associated Press, October 2026] [Pulse 2]. Public materials also say the platform can test domains, applications, IP ranges, cloud accounts, and internal networks, and supports scheduled assessments, on-demand testing, retesting, and reporting aligned with common security and compliance frameworks [Associated Press, October 2026].
The operating model appears designed to reduce setup friction. Citybiz and Pulse 2 both reported that external assessments require no software installation and that customers can authorize a target domain and receive initial results the same day, although that speed claim should still be read as launch-stage reporting rather than a broad service-level record [Citybiz, October 2026] [Pulse 2]. Pricing starts at $249 per month according to the company website, which is directionally important for positioning but remains company-supplied evidence rather than independently corroborated commercial data [avalonx.ai, retrieved 2026].
AvalonX attributes the system to a proprietary model it calls Aether. Press coverage and company materials describe Aether as a 400-billion-parameter cybersecurity AI trained on nine years of real-world offensive-security data and supported by 60 billion dark-web records [Associated Press, October 2026] [Pulse 2] [avalonx.ai, retrieved 2026]. Those claims help explain the company's differentiation story, but the public record does not yet include a verified demo, benchmark, or third-party evaluation of model performance, false-positive rates, or remediation accuracy.
Partially corroborated -- Product scope is corroborated by Associated Press and Citybiz, while pricing and model-detail claims rely in part on company materials and publisher reprints.
Market Research
PUBLIC
The market matters now because AvalonX is selling into two forces that are already in motion, small and midsize businesses facing enterprise-grade attack pressure, and a security-buying base that still struggles to afford or operationalize traditional penetration testing [Associated Press, October 2026] [Citybiz, October 2026].
The public record here is thinner on formal market sizing than on the product claim itself. No named third-party TAM, SAM, or SOM study was captured in the source set, so any hard sizing exercise would be speculative. What is observable is the wedge AvalonX has chosen: continuous, subscription-based penetration testing for small and midsize businesses, with launch pricing that starts at $249 per month according to the company website [Associated Press, October 2026] [avalonx.ai]. That framing places the company at the intersection of offensive security services, external attack-surface monitoring, and security tooling sold downmarket to organizations that may not maintain large in-house teams [Associated Press, October 2026] [Citybiz, October 2026].
The demand signal in the cited coverage is practical rather than statistical. Both launch reports describe a product built for buyers that want faster testing cycles than periodic pen tests can usually provide, and that may need lower-friction deployment, same-day initial results, and reporting mapped to common compliance frameworks [Associated Press, October 2026] [Citybiz, October 2026]. If those claims hold in production, the appeal is straightforward: reduce the gap between discovering an exposure and validating it, without asking an SMB to staff a mature offensive-security function internally [Associated Press, October 2026] [Pulse 2].
Adjacent markets matter because AvalonX is not entering a clean, isolated category. The company touches at least four neighboring budgets: conventional penetration testing services, vulnerability management, attack-surface management, and broader compliance-oriented security tooling [Associated Press, October 2026] [Citybiz, October 2026]. That can expand the addressable spend if buyers treat continuous testing as a replacement for point-in-time engagements, but it can also lengthen sales cycles if the product is evaluated against several existing tools or outside consultants rather than against a single line item.
| Market lens | Publicly supported reading | Source basis |
|---|---|---|
| Core market | Continuous, AI-powered penetration testing for SMBs | [Associated Press, October 2026] |
| Adjacent budget | Traditional penetration testing services | [Associated Press, October 2026] [Citybiz, October 2026] |
| Adjacent budget | External attack-surface and vulnerability monitoring | [Citybiz, October 2026] [Pulse 2] |
| Adjacent budget | Security and compliance reporting workflows | [Associated Press, October 2026] |
The table shows a company framed less as a pure-service consultancy and more as a software layer trying to absorb several security workflows at once. That is strategically attractive if buyers consolidate spend, but it also means the relevant market benchmark is broader than pen testing alone.
Regulatory and macro support are implied more than directly documented in the captured sources. The product's emphasis on scheduled assessments, retesting, and reporting aligned with common security and compliance frameworks suggests that audit readiness and customer-security expectations are part of the buying motion, especially for SMBs selling into larger enterprises [Associated Press, October 2026]. At the same time, the available public evidence does not establish which regulations or industry standards are currently driving demand, so the safer reading is that AvalonX is positioned to benefit from a general shift toward continuous validation, not from any one disclosed mandate.
Partially corroborated -- Market characterization is supported by Associated Press and Citybiz, but no independent third-party TAM or growth study was captured in the public source set.
Competition and Substitutes
MIXED AvalonX is positioning itself between traditional point-in-time penetration testing and enterprise-heavy attack-surface platforms, with a stated focus on continuous testing for SMBs at entry pricing that begins at $249 per month according to the company website [Associated Press, October 2026] [avalonx.ai].
| Company | Positioning | Stage / Funding | Notable Differentiator | Source |
|---|---|---|---|---|
| AvalonX | Continuous, AI-powered penetration testing for small and midsize businesses [PUBLIC] | Seed, up to $10 million from Ridown Capital announced in October 2026 [PUBLIC] | Artemis combines autonomous testing with human tester verification; AvalonX says pricing starts at $249 per month and that Aether is trained on nine years of offensive-security data supported by 60 billion dark-web records [PUBLIC] | [Associated Press, October 2026]; [avalonx.ai]; [Citybiz, October 2026] |
| Cobalt | Penetration testing platform and pentest-as-a-service provider [PUBLIC] | Not disclosed here | Human-led platform model is the clearest named alternative in the available fact set [PUBLIC] | [STRUCTURED FACTS competitor set, retrieved 2026] |
The competitive map is still narrow in the public record, which matters because AvalonX launched only in October 2026 after 18 months in stealth [Associated Press, October 2026]. On one side sit incumbent penetration-testing firms that sell periodic assessments, usually human-led and often priced for larger enterprise budgets rather than SMB subscriptions [Associated Press, October 2026]. Adjacent substitutes include internal security teams and compliance-driven vulnerability management workflows, especially for smaller companies that may choose cheaper scanning tools over recurring offensive testing, an adoption risk implied by AvalonX's own SMB affordability pitch [Associated Press, October 2026] [avalonx.ai].
Where AvalonX appears to have an edge today is product packaging rather than proven distribution. The company is presenting a simpler buying motion for SMBs, domain input, no required software installation for external assessments, and same-day initial results according to Citybiz and Pulse 2, although these operational claims remain lightly corroborated [Citybiz, October 2026] [Pulse 2]. The more substantive differentiator, if accurate, is the data and model layer: AvalonX says Aether is trained on nine years of real-world offensive-security data and supported by 60 billion dark-web records, while Artemis uses autonomous agents to simulate attacker behavior and then routes findings to human testers for verification [Associated Press, October 2026] [Pulse 2]. That edge is potentially durable only if the underlying data corpus is difficult for peers to replicate and if the human-review layer keeps false positives low enough for small security teams to trust the output. At this stage, both conditions are asserted more than demonstrated in public.
The exposure is easier to see than the moat. Cobalt's specific advantage, based on category position alone, is that it is already known as a penetration-testing platform, while AvalonX is just entering the market with no named customers, no public deployment references, and no public evidence yet of channel ownership or enterprise procurement traction [Associated Press, October 2026]. AvalonX is also exposed to categories above its current lane. The available material frames the product around SMB accessibility and affordability, not deeply regulated large-enterprise environments with long buying cycles and formal red-team requirements [Associated Press, October 2026] [avalonx.ai]. If buyers decide that human-led assurance is still necessary for higher-stakes assessments, then an automation-first posture could narrow AvalonX's addressable share inside premium pentest budgets even if it expands access at the low end.
The most plausible 18-month scenario is a split market rather than a winner-take-all outcome. Winner if execution goes right: AvalonX, if it can convert its low-friction subscription model into repeat SMB adoption and prove that autonomous testing plus human verification produces useful remediation without enterprise-level spend [Associated Press, October 2026] [Citybiz, October 2026]. On the evidence available publicly, the competitive question is less whether there is room for another security vendor and more whether AvalonX can turn a credible product narrative into a repeatable channel before better-known testing providers respond.
Reasoned from indirect evidence -- This section relies on solid public sourcing for AvalonX's positioning and launch details from Associated Press and Citybiz, but the named competitor set is limited to one structured-fact competitor and several differentiators remain company-asserted or lightly corroborated.
Opportunity
PUBLIC
The prize here is unusually large for a company this early: if AvalonX can make continuous penetration testing cheap enough, fast enough, and credible enough for smaller enterprises, it has a path to becoming the default offensive-security layer for a part of the market that has historically been priced into periodic, consultant-led work [Associated Press, October 2026] [Citybiz, October 2026].
The headline opportunity rests on a simple shift in buying behavior. AvalonX is not presenting itself as a point-in-time pen test firm; it is positioning Artemis as a subscription product for continuous testing, with same-day initial results for external assessments, no software installation required in that use case, and human testers included for verification of findings [Associated Press, October 2026] [Citybiz, October 2026] [Pulse 2]. That combination matters because the SMB and midmarket security buyer usually lacks the staff and budget for frequent bespoke red-team work, but still needs evidence for exposure management and compliance workflows [Associated Press, October 2026]. If AvalonX can reliably convert a traditionally episodic service into a recurring software budget line, the reachable outcome is not merely a successful security tool vendor. It is a platform that normalizes continuous offensive testing as routine infrastructure for smaller organizations.
The upside paths are still narrow, but they are legible from the public record.
| Scenario | What happens | Catalyst | Why it's plausible |
|---|---|---|---|
| SMB security standard | AvalonX becomes the default continuous pen-testing subscription for SMBs and upper midmarket companies that cannot support in-house offensive security teams. | The October 2026 launch couples a subscription model with entry pricing from $249 per month and same-day initial external results, lowering procurement and deployment friction [avalonx.ai] [Citybiz, October 2026]. | The company is explicitly targeting small and midsize businesses with continuous testing rather than one-off projects, and the product spans domains, applications, IP ranges, cloud accounts, and internal networks [Associated Press, October 2026]. |
| Compliance-adjacent control layer | Artemis becomes a recurring system of record for customers that need scheduled testing, retesting, and framework-aligned reporting. | Broader adoption of security and compliance programs that require repeatable evidence could favor products that automate assessments and retesting cycles [Associated Press, October 2026]. | Public materials already describe scheduled and on-demand assessments, retesting, and reporting aligned with common security and compliance frameworks, which is closer to an operating workflow than a one-time scan [Associated Press, October 2026]. |
| Data-advantaged offensive-security engine | AvalonX uses usage and testing data to improve attack simulation and remediation quality faster than smaller services firms or simpler automation vendors. | Product usage at launch feeds back into Aether and Artemis, strengthening detection, attack-path mapping, and prioritization over time, if customers continue running assessments [Associated Press, October 2026] [avalonx.ai]. | AvalonX says Aether is trained on nine years of real-world offensive-security data and supported by 60 billion dark-web records, while Artemis uses autonomous agents to simulate attacker behavior and map movement through systems [Associated Press, October 2026]. |
The compounding mechanism, if it works, is data plus workflow plus trust. Each additional customer assessment can in theory sharpen the platform's understanding of attacker behavior, while the inclusion of human verification gives AvalonX a better chance of keeping false positives low enough for smaller teams to act on the output [Associated Press, October 2026] [Pulse 2]. At the same time, scheduled assessments, on-demand testing, retesting, and reporting create recurring touchpoints that can turn a low-entry-price purchase into a broader security operating habit [Associated Press, October 2026] [avalonx.ai]. The public evidence does not yet show that this flywheel is already producing measurable commercial traction, but the product design is at least consistent with a software compounding model rather than a pure labor model.
Sizing the win is harder because the current public record does not include a verified market sizing dataset or a clean public comparable for this exact product configuration. Still, the funding announcement frames AvalonX as pursuing a software model in cybersecurity, not a boutique consultancy, and the platform claim spans external attack surface, applications, cloud accounts, and internal networks [Associated Press, October 2026]. If the "SMB security standard" scenario took hold, the logical end state would be a scaled security SaaS company with recurring subscription revenue and a data asset built from continuous offensive testing. That could support a multibillion-dollar outcome (scenario, not a forecast), but the public evidence today is not strong enough to anchor that figure to a named comparable without overstating precision. At this stage, the more defensible observation is that AvalonX is aiming at a category large enough to matter, with a product thesis that could expand from a narrow penetration-testing wedge into a broader offensive-security platform if customers adopt it as ongoing infrastructure [Associated Press, October 2026] [Citybiz, October 2026].
Partially corroborated -- Based primarily on Associated Press launch coverage, with partial corroboration from Citybiz and company materials; upside sizing remains directional because no independent market dataset or public traction benchmarks were confirmed.
Sources
Open sources
[Associated Press, October 2026] AI STARTUP AVALONX EMERGES FROM STEALTH WITH $10 MILLION IN FUNDING FROM RIDOWN CAPITAL | https://apnews.com/press-release/getnews/press-release-bf5cc51d4015fb7a7c13fe17c5ee57c7
[Citybiz, October 2026] AvalonX Launches AI-Powered Penetration Testing Platform With Up to $10M in Funding | https://www.citybiz.co/article/927074/avalonx-launches-ai-powered-penetration-testing-platform-with-up-to-10m-in-funding/
[Pulse 2] AvalonX Emerges From Stealth With Up To $10 Million In Funding To Solve SMB Cybersecurity | https://pulse2.com/avalonx-emerges-from-stealth-with-up-to-10-million-in-funding-to-solve-smb-cybersecurity/
[avalonx.ai] AvalonX | https://www.avalonx.ai/
Articles about AvalonX
- AvalonX Puts a Human Pen Tester Inside Its $249 AI Subscription — The startup's Artemis platform, powered by a proprietary AI model, aims to make continuous security testing accessible for smaller businesses.