Cyber Security News

AkiraBot Floods 80,000 Sites After Outsmarting CAPTCHAs and Slipping Past Network Defenses

AkiraBot, identified by SentinelLABS, represents a sophisticated spam bot framework that targets website chats and contact forms to promote low-quality SEO services.

Since its inception in September 2024, AkiraBot has impacted over 420,000 unique domains, successfully spamming at least 80,000 websites.

It leverages both CAPTCHA evasion techniques and network detection evasion to elude website security measures, employing OpenAI’s services for generating custom spam messages tailored to the content of the targeted websites.

AkiraBot’s origins trace back to late 2024 when it was first identified targeting Shopify websites.

Over time, it evolved to target multiple website platforms like GoDaddy, Wix, and Squarespace, which are popular among small and medium-sized businesses for their integration capabilities with eCommerce, content management, and business services.

CAPTCHA Evasion and AI-Generated Spam Messages

The bot uses multiple CAPTCHA bypass services such as Capsolver, FastCaptcha, and NextCaptcha, functioning as failovers when direct browser emulation fails.

AkiraBot’s core functionality includes a sophisticated mechanism for generating unique spam content.

It scrapes the targeted website’s HTML, processes it through BeautifulSoup, and then sends a prompt to OpenAI’s gpt-4o-mini model, instructing it to tailor outreach messages that are contextually relevant.

This adaptive spam approach complicates traditional spam filtering efforts by rotating the supplied domains, supplying tailored messages unique to each interaction, and employing services like Reamaze for spamming chat widgets.

AkiraBot’s evasion capabilities extend beyond CAPTCHA bypass to network detection evasion.

The use of SmartProxy, advertised as ethically sourced, has become a point of concern, as it garners interest from cybercriminals.

SmartProxy credentials from BlackBasta leaks

This service allows AkiraBot to rotate through a diverse array of IP addresses, making it challenging to block the bot’s traffic effectively.

The Bot’s Infrastructure: Domains and Operators

The bot’s infrastructure leverages two branding themes, ‘Akira’ and ‘ServiceWrap’, in its domain naming convention.

Older domains like akirateam[.]com and goservicewrap[.]com were used to establish a presence, while newer iterations focus on evading detection through continuous domain and IP address rotation.

SentinelLABS identified consistent proxy credentials and test sites across different archives, suggesting that the same group or individual operates multiple versions of the bot.

AkiraBot’s success across different website platforms illustrates its flexibility and the ongoing battle against it.

AkiraBot GUI

Shopify, GoDaddy, Wix, and Squarespace have all implemented security measures to combat such spam, yet AkiraBot continues to adapt.

OpenAI’s investigation into the misuse of their services indicates a commitment to disabling associated API keys and assets involved in such illicit activities.

Detecting and mitigating AkiraBot presents multiple challenges. Its use of LLM-generated content means each message is unique, reducing the effectiveness of signature-based filtering.

The dynamic rotation of domains and the bot’s ability to efficiently bypass CAPTCHA controllers demand innovative security solutions.

Website hosting providers are compelled to continually update their defenses, often resulting in an arms race with cybercriminal actors.

AkiraBot’s campaign against websites underscores the evolving threat faced by online service providers.

Its capability to generate unique spam content and evade detection through multiple methods highlights the growing sophistication in bot frameworks.

This persistent challenge requires a collaborative approach between hosting providers, AI service providers, and cybersecurity researchers to curb AkiraBot’s proliferation.

Indicators of Compromise

To aid in the detection of AkiraBot activities, SentinelLABS has identified several domains and associated SHA-1 archive hashes:

  • Domains: akirateam[.]com, beservicewrap[.]pro, go-servicewrap[.]com, etc.
  • Archives SHA-1: 09ec44b6d3555a0397142b4308825483b479bf5a, 0de065d58b367ffb28ce53bc1dc023f95a6d0b89, etc.

AkiraBot’s ability to adapt and outmaneuver multiple security layers has led to widespread compromise of websites, necessitating a unified response to address and mitigate this emerging threat landscape.

Find this News Interesting! Follow us on Google NewsLinkedIn, & X to Get Instant Updates!

Aman Mishra

Aman Mishra is a Security and privacy Reporter covering various data breach, cyber crime, malware, & vulnerability.

Recent Posts

Microsoft Defender XDR False Positive Leaked Massive 1,700+ Sensitive Documents to Publish

An alarming data leak involving Microsoft Defender XDR has exposed more than 1,700 sensitive documents…

2 minutes ago

‘SessionShark’ – A New Toolkit Bypasses Microsoft Office 365 MFA Security

Security researchers have uncovered a new and sophisticated threat to Microsoft Office 365 users: a…

46 minutes ago

Hackers Exploit MS-SQL Servers to Deploy Ammyy Admin for Remote Access

A sophisticated cyberattack campaign has surfaced, targeting poorly managed Microsoft SQL (MS-SQL) servers to deploy…

1 hour ago

New Report Reveals How AI is Rapidly Enhancing Phishing Attack Precision

The Zscaler ThreatLabz 2025 Phishing Report unveils the alarming sophistication of modern phishing attacks, driven…

1 hour ago

159 CVEs Exploited in the Wild in Q1 2025, 8.3% Targeted Within 1-Day Vulnerabilities Exploited

VulnCheck's latest report for Q1 2025 has identified 159 Common Vulnerabilities and Exposures (CVEs) publicly…

1 hour ago

Hackers Claim TikTok Breach, Leak Over 900,000 Usernames and Passwords

A hacker collective known as R00TK1T claims to have breached TikTok's user database, allegedly leaking…

2 hours ago