Credential stuffing assaults had a huge effect in 2024, fueled by a vicious circle of infostealer infections and knowledge breaches. But issues might be about to worsen nonetheless with Computer-Using Agents, a brand new form of AI agent that allows low-cost, low-effort automation of widespread net duties — together with these ceaselessly carried out by attackers.
Stolen credentials: The cyber prison’s weapon of alternative in 2024
Stolen credentials have been the #1 attacker motion in 2023/24, and the breach vector for 80% of net app assaults. Not stunning when you think about the truth that billions of leaked credentials are in circulation on-line, and attackers can choose up the newest drop for as little as $10 on prison boards.
The prison market for stolen credentials is benefitting from the publicity of high-profile breaches in 2024 such because the assaults on Snowflake prospects utilizing credentials present in knowledge breach dumps and compromised credential feeds from infostealer and mass phishing campaigns, ensuing within the compromise of 165 buyer tenants and tons of of thousands and thousands of breached information.
But regardless of 2024 being an unprecedented yr by way of the affect of identity-based assaults, there’s nonetheless a variety of unfulfilled potential for attackers to comprehend.
Credential assault automation — what’s modified with the shift to SaaS?
Brute forcing and credential stuffing are nothing new, and have been a key element of the cyber attacker toolkit for many years. But it is not fairly as straightforward to routinely spray credentials throughout techniques because it as soon as was.
No extra one-size-fits-all
Rather than a single centralized community with apps and knowledge contained inside an infrastructure perimeter, enterprise IT is now fashioned of tons of of web-based apps and platforms, creating 1000’s of identities per group.
This implies that identities too are actually decentralized and distributed all around the web, versus being saved solely in identification techniques like Active Directory, and applied utilizing widespread protocols and mechanisms.
While HTTP(S) is normal, trendy net apps are advanced and extremely personalized, with a graphically-driven interface that’s totally different each time. And to make issues worse, trendy net apps are particularly designed to stop malicious automation by way of bot protections like CAPTCHA.
So relatively than encountering normal protocols and having the ability to write a single set of instruments to make use of throughout any group/atmosphere e.g. write a DNS scanner as soon as, use a single port scanner like Nmap for your complete web, write a single script per service (e.g. FTP, SSH, Telnet, and so on.) on your password sprayer — customized device improvement is as an alternative required for each app that you just need to goal.
Finding the needle within the haystack
Not solely are there extra environments for attackers to incorporate within the scope of their assault, however there are extra credentials to work with.
There are round 15 billion compromised credentials obtainable on the general public web, not together with these discovered solely in personal channels/feeds. This record is rising the entire time — like 244M never-before-seen passwords and 493M distinctive web site and e mail tackle pairs being added to Have I Been Pwned from infostealer logs simply final month.
This sounds scary, however it’s tough for attackers to harness this knowledge. The overwhelming majority of those credentials are outdated and invalid. A latest evaluation of TI knowledge by Push Security researchers discovered that fewer than 1% of stolen credentials included in menace intelligence feeds from a multi-vendor knowledge set was actionable — in different phrases, 99% of compromised credentials have been false positives.
But not all of them are ineffective — because the Snowflake assaults demonstrated, which efficiently leveraged credentials courting again to 2020. So there are clearly treasures ready to be found by attackers.
Attackers are pressured to prioritize
The distributed nature of apps and identities, and the low reliability of compromised credential knowledge, means attackers are pressured to prioritize — regardless of a target-rich atmosphere of tons of of enterprise apps, creating 1000’s of sprawled identities per group, as a result of:
- Writing and operating customized python scripts for each single app (there are greater than 40k SaaS apps on the web) will not be reasonable. Even should you did the highest 100 or 1000 that might be a big activity and require fixed upkeep, whereas barely scratching the floor of the full alternative.
- Even when totally scripted and utilizing a botnet to distribute the assault and keep away from IP blocking, controls like price limiting, CAPTCHA, and account lockouts can impede mass credential stuffing towards a single app. And a concentrated assault on a single web site goes to generate vital ranges of site visitors if you wish to get by way of 15 billion passwords in an inexpensive timeframe, so it is very prone to increase the alarm.
So attackers have a tendency to focus on a smaller variety of apps, and solely search for a direct match by way of the credentials tried (e.g. the stolen credential should straight belong to an account on the goal app). When they do go after one thing new, it tends to be targeting a selected app/platform (e.g. Snowflake) or on the lookout for a narrower subset of credentials (e.g. credentials clearly related to edge gadgets, for extra conventional community environments).
A missed alternative?
As we have established, the state of affairs concerning credential stuffing assaults is already fairly dangerous regardless of these limitations. But issues might be considerably worse.
Password reuse means a single compromised account might flip into many
If attackers have been capable of improve the dimensions of their assaults to focus on a broader variety of apps (relatively than concentrating on a shortlist of excessive worth apps) they might benefit from all-too-common password reuse. According to a latest investigation of identification knowledge, on common:
- 1 in 3 workers reuse passwords
- 9% of identities have a reused password AND no MFA
- 10% of IdP accounts (used for SSO) have a non-unique password
What does this imply? If a stolen credential is legitimate, there is a good probability that it may be used to entry a couple of account, on a couple of app (at the least).
Picture the situation: A latest compromised credential leak from infostealer infections or credential phishing campaigns exhibits {that a} explicit username and password mixture is legitimate on a selected app — for instance Microsoft 365. Now, this account is fairly locked down — not solely does it have MFA, however there are conditional entry insurance policies in place proscribing the IP/location it may be accessed from.
Usually, that is the place the assault would finish, and also you’d flip your consideration to one thing else. But what should you have been capable of spray these credentials throughout each different enterprise app that the person has an account on?
Scaling credential assaults with Computer-Using Agents
Until now, the affect of AI on identification assaults has been restricted to the usage of LLMs for the creation of phishing emails, in AI-assisted malware improvement, and for social media bots — little question vital, however not precisely transformative, and requiring fixed human oversight and enter.
But with the launch of OpenAI Operator, a brand new form of “Computer-Using Agent”, this might be about to vary.
Operator is educated on a specialist dataset and applied in its personal sandboxed browser, that means it is ready to carry out widespread net duties like a human — seeing and interacting with pages as a human would.
Unlike different automated options, Operator requires no customized implementation or coding to have the ability to work together with new websites, making it a way more scalable possibility for attackers trying to goal a broad sweep of web sites/apps.
Demo: Using Operator to conduct credential stuffing assaults at-scale
Researchers at Push Security put the malicious use-cases of Operator to the take a look at, utilizing it to:
- Identify which corporations have an current tenant on an inventory of apps
- Attempt to login to varied app tenants with a supplied username and password
Impact abstract
The outcomes have been fairly eye-opening. The operator clearly demonstrated the flexibility to focus on an inventory of apps with compromised credentials and carry out in-app actions. Now take into consideration this x10, x100, x10,000 … These are usually not advanced duties. But the worth of CUAs Operator will not be in tackling complexity, however scale. Imagine a world the place you may orchestrate Operator home windows by way of API and get it to execute these actions concurrently (performance that exists already for ChatGPT).
But that is larger than Operator — it is in regards to the route of the expertise. OpenAI might implement restrictions — higher in-app guardrails, price limits on the variety of concurrent duties and whole utilization, and so on. But you may assure it will not be the one CUA — it is solely a matter of time earlier than comparable merchandise emerge (perhaps even inherently malicious ones) making use of the identical expertise.
Final ideas
It’s nonetheless early days for CUA tech, however there is a clear indication that an already extreme safety problem might be made worse with this explicit type of AI-driven automation. While the flexibility to focus on a broad set of apps has been beforehand past the scope of conventional automation, it is about to develop into far more accessible to even low-skilled attackers (suppose: subsequent gen script kiddies?).
Another manner to consider it’s that it successfully offers a human attacker a fleet of low-level interns who do not fairly know what they’re doing, however could be instructed to carry out particular, itemised duties at scale with solely the occasional examine in — when you work on different, extra advanced duties. So, a bit like a pink workforce supervisor of AI bots.
Operator implies that attackers can leverage compromised credentials at-scale, benefit from the huge numbers of susceptible and misconfigured identities, and convert them into systemic breaches far more simply. In a manner, it might make credential stuffing a bit extra prefer it was earlier than the shift to cloud apps — the place you may spray 1000’s of credentials throughout your targets while not having customized improvement each time.
Thankfully, no new anti-AI capabilities are required — however it’s extra essential than ever that organizations look to defend their identification assault floor and discover and repair identification vulnerabilities earlier than attackers can benefit from them.
Find out extra
If you need to be taught extra about identification assaults and tips on how to cease them, try Push Security — you may e-book a demo or check out their browser-based platform at no cost.
And if you wish to see them demo extra malicious use circumstances of Operator, try this on-demand webinar.