The whole concept of Google Tag Manager was that you could give marketing departments the ability to add "tags" to the website without having to involve engineering. That's how we got 10MB of javascript with every news article.
Client side A/B testing through services like Optimizely and VWO and Adobe Target is a massive security nightmare in of itself. You're giving anyone with access to a third party service the ability to run any code they like on your website, and change any and all interface elements as they see fit.
Even one of these accounts being compromised would be a massive security incident that would potentially put millions of people's data at risk.
Just pointing out that these testing solutions can also be deployed server-side instead of via client-side tags. I suspect that is the route a lot of their customers will take now.
It also makes adding csp a nightmare. Your marketing department is putting in 30 trackers, changing monthly. And then there’s different countries for end clients with different TLDs for connect-src. You can’t possibly get anything locked down enough. That’s bad for security when you rely on csp for mitigating other factors.
I'm sure they know how to serve from a single domain - but it is in their interest if you are forced to give them the nonce pass. So yeah, it's (again) all about incentives.
Client side security isn't the issue, but CSP is kind of a shit show. The configuration apparatus is godawful. I think the best way to think of it is if someone turned CSS (yes, the styling language) into a security configuration protocol. It uses the same style of cascading overrides that CSS employs.
If it's available, pull the third party JS in via package manager rather than a CDN you don't control.
NPM isn't perfect but there's typically a substantial delay between the release of a malicious package and that code getting onto your production environment. Unless you're automatically updating every day, there's a good chance someone will discover the problem before it affects you. And you can always scan your upstream dependencies.
This doesn't work if you're using someone else's CDN. Any change to their content is distributed to all your users immediately.
A certificate-based approach for scripts residing on different domains, for example, so you can be sure it hasn't been tampered with.
A javascript file served from example-a.com should be able to run on example-b.org only if it has been signed with the same TLS key used for example-a.com .
CSP allows you to configure your loaded JS with a complementary hash check. If you're the one who puts the script on your page, in theory, it should be safe for as long as remains the same.
Vet everything, cut down to 2-3 services tops. Selfhost analytics whenever possible. Point to the OP article when asked why we can’t blindly put 30 tags using GTM.
Then you've done your job and move on. Just make sure that there's enough public record of your warning that - when the next cdn supply chain attack happens - you can say "I warned you about this and you decided not to take action".
No organization is perfect. I don't think this is something to quit over. Just make sure to CYA.
That is not any more secure if you are just blindly proxying the third party scripts and in fact is worse since it runs in the context of your own domain.
I mean, really that situation is CSP finally earning its keep. Put whatever you want in tag manager but at the end of the day it's not going to work unless I review it.
Not going to happen. Most decisions in tech do not factor in risk, only cost. Risk mitigation usually requires considerable spend, and very few people/companies are willing to spend on low-probability events.
I know you're sort of speaking abstractly, and I think you're correct that this is how companies interpret risk, but it's interesting to give a moment's thought to -- Murphy's Law tells us all events have a 100% probability of occurring given a certain amount of time. And we tend to see that proven out time and again.
This might change as AI-assisted hacks and outages become more frequent. A materialized risk with associated costs is very different from an abstract risk that's hard to quantify.
There were a couple of years before the concept of tag manager where publishers would just blindly call a JS file on their trafficker's server. This file was maintained by engineering, and you would still end up with MB of junk on every page (especially since this was before any kind of tracking regulation, wild things would happen behind the scenes). So this is more on the advertising industry itself than on wannabe devs PMs.
Beyond all the issues with third party JavaScript inclusion on sites blowing up their security perimeter, there are also significant issues with the link tracking for these as they nest and re-encode their values so much in the URLs any reasonable firewall concludes that they're some kind of obfuscation attack.
What is that saying? A person will not understand something that their paycheck requires them to not understand.
That’s why it needs to become criminal for a website to serve malware, including through third party channels like ad networks and analytics services. Serving shit websites is no different than knowingly, or negligently, dumping toxic waste and should be treated the same.
If it's criminal for sites to accidentally serve malware, what does it mean for open source projects to accidentally get hit by supply chain attacks, or for them to get compromised by a rogue contributor (eg. xz)?
Those website operators include third-party code willingly and explicitly. There was never an accident. There is no supply-chain attack here: you cannot seriously claim that unaudited, unauthenticated terminal with root security principal getting used is an attack.
If your website is making any requests outside of the original domain, there's no "accidental" in anything the website does.
A business in the US has legal liability if they fail to do their due diligence in their supply chain and it turns out their suppliers are involved in human trafficking, child labor, forced labor, etc. or is using hazardous materials in the product.
There should be standards that people have to live up to if they are going to provide a product, even for free, to others.
One could distinguish between criminal negligence and forgivable negligence in court. Unfortunately for GP that probably means you can't actually stick charges on anybody who serves malware via an ad network like this.
Easy solution: Don't use ad networks. That'll likely free your site from illegal tracking, as well as ad trash, and will even improve your lIgHtHoUsE score!
> A person will not understand something that their paycheck requires them to not understand.
I never really understood this idiom, because how can a PHB engineering manager (for example) be required to not understand their team's technology in order to retain their job? Surely if that manager knows their team's work inside out then they would be a more effective manager - which is good for everyone?
If I squint-my-brain, I can see how (for example) that anyone who works full-time as an SEO "consultant" in the year 2026 would only be if they actually believed that SEO tricks actually worked, which requires them to not understand how modern search-engines' spiders actually work (because it's my belief that "SEO" has been either nonsense bollocks - or a complete grift - since the mid-2010s). Am I close?
Because it is not the PHB that starts the problem. It is someone from some other department (usually marketing) who wants tracking, wants it now and wants it done with GTM and will whine to th C-levels until the engineers add Google Tag Manager.
It's not the team's technology that's being referred to. It's about hazards. People whose job it is to do X have a hard time understanding that X shouldn't be done.
In this case, it's the dangers of the tracking and advertising frameworks being used for hostile purposes.
> because it's my belief that "SEO" has been either nonsense bollocks - or a complete grift - since the mid-2010s
It's a continuous case of whack-a-mole. SEO consultants think <thing> would be a good way to improve the search ranking of their customers, it works, it spreads throughout the industry, Google catches on, rinse and repeat.
As soon as something is a KPI, there is incentive to game the system - with the top end of "gaming the system" toeing or crossing the line into fraudulent, anti-competitive or otherwise illegal territory.
"There are none so blind as those who will not see" - John Heywood (et al.)
The follow on line from that one is even better: "The most deluded people are those who choose to ignore what they already know". I would probably want to add on things like "because it's not convenient" or similar for particular relevance to some of the stuff I've been dealing with recently at work. Folks know it's dangerous to do something a certain way, and know how to do it a safe way, but it's not as convenient so they delude themselves it'll be fine just this once. And just this once too. And just this once, once more. Slowly but surely normalising the deviance.
In the context of the Optimizely case, people know they're giving it dangerous levels of capabilities, but it's just too darn inconvenient to do it in a safe way so sure, why not. It'll be fine (and it will be, right up until it isn't and you realise your choice has compromised your users)
Just had marketing try to add all OTHER previous page events to a new event, because they didn't want to to go through the trouble of creating a dashboard using the user funnel to this feature.
They knew they could do it, and how, they just didn't want to spend 10 minutes doing it themselves. To them, asking an engineer to do it in code is easier because it only takes a minute to ask.
Marketing folks they were the driving force behind Adobe Flash, but the issues with that were so great that the industry did eventually get rid of it. It just takes time for the costs to become sufficiently apparent to everyone.
AI-assisted hacks and outages are accelerating the process of making the costs of weak security apparent.
The owners of the ad platforms end up dictating what data is used, they have the leverage and also marketing departments with power end up forcing engineering to use the audience tools Google and Meta provide. At least GTM gets blocked by ad-blockers.
We got 10MB of javascript (maybe 10 years ago...) because engineering needs to build one frontend with multi browser and multi screen support so they picked an off-the-shelf do-everything framework like Amber or Angular, or a react/redux build with 300 nested dependencies.
No, the 10MB (sometimes way more) of JavaScript for a damn news article has nothing to do with SPA frameworks (Angular) or libraries like React. It is tracking malware put there by marketing/advertising idiots.
I remember when this first became a thing when page size was important. I fought tooth and nail with marketing to try and block tag manager. I've had third party URLs block production apps in Chrome before because some marketing url was hosting malicious code.
The most infuriating thing, when you do remove these tags, marketing forgets they exist and are like "Oh we don't use that anymore".
Porqué no los dos? Both are true. Heavy JS frameworks, ads, trackers, etc., but also heavy web fonts, of which maybe 5 icons are used for the site, but the font weighs a megabyte. Just in case you need it.
Those were decisions at a time engineering was time constrained. All those ancillary things are returning to engineering since its just one more prompt now to manage it.
This stuff is so avoidable but unless we implement meaningful reprocussions things aren't going to change. There is no accountability.
In this case optimizely should have had better systems to spot a release like this (internal cool downs, check and balances, phased rolouts, automatic security checks). And sites that have massive user bases (cnn, bbc, etc..) have zero excuse to not have the same.
I would go so far as to say those sites should not be serving from vendors but should be vendoring their own scripts that they also are passing through security audits and serving them selves with cool downs.
How do these companies not have better dedication to protecting their users is beyond me.
Are you sure? The affected companies are going to totally make Optimizely pay for damages. That could become a lot of money pretty fast. And tbh, I'm not sure that the average ToS's maximum liability ceiling will hold up in court with a blast radius this big.
Usually, the plaintiff/claimant has to prove the damages they suffered, not the damages their (non-paying) users suffered. So then they would have to claim reputation damage or something like that, and that might not be so easy to prove.
It could happen but if you look at the last few decades of losses from software bugs, it’s exceedingly rare for the cost to be more than lost future business, and even that’s unusual. Remember when CrowdStrike brought down businesses around the world and they suffered … having their share price more than double?
Accountability is incompatible with tech sector as it currently exists. The entire ethos of move fast and break things is incompatible with it never mind the shady adtech part of the sector.
Partially but don’t think that quite follows. A lot of businesses aren’t as fraught with conflicts of interest as tech.
A company making say bricks has an incentive to cut costs too certainly but they’re not implementing loot boxes, tracking people semi covertly, implementing dark patterns to boost engagement by any means or heaven forbid launching missies based on sketchy AI decisions.
> I would go so far as to say those sites should not be serving from vendors but should be vendoring their own scripts that they also are passing through security audits and serving them selves with cool downs.
This is the bare minimum. Automatic 3rd party requests should be considered data unsolicited data disclosure under GDPR.
There's a missing step in the explanation. Do major websites like CNN and BBC handle cryptocurrency or credit cards on the same website? Why? Or is there also a browser security bug?
I had the same question, and I really wish these reports would make in clearer what exactly the malware can and can't do. The whole point of web browsers is that sites run in a sandbox so that even if a site does serve malicious code, it should be limited to just that site.
From what I could tell, the malware just generally looked at anything on the page, in cookies/local storage, or clipboard that looks like a crypto address or credit card number. So for sites like CNN or BBC they shouldn't be able to do much damage unless there is a subscription page for example that takes credit card number.
However, the malware also delivered the Coruna and DarkSword iOS exploit kits, which do take advantage of browser vulnerabilities in iOS to cause more serious damage.
I feel that whole web site model is broken. They are run in "sandbox" same way AI companies run their sandboxes. Not caring at all where they do call... So every other site needs to implement protections against other sites.
The model is broken on fundamental level and it will never be fixed.
I imagine CNN does, but they might have been smart enough to unburden the payment pages from the marketing/surveillance cruft. Or perhaps not. There have also been browser escape vulnerabilities.
I've never understood why Optimizely is a thing. Why is A/B testing so hard? I implemented a full A/B test system for a major media website in 1 day about 10 years ago and its still in use today (I can pull the the secret report on any page and I just did). Its just simple induction algorithms, tagging and conditional rendering.
Marketing teams don't have engineering resources, generally speaking. But they do have the clout to get engineering to install this little tool that will help them "self service."
This is likely a result of Optimizely cutting headcount on their legacy product engineering teams to reallocate funding towards their AI product. We’re going to see more stories like this for every company taking a similar approach.
I guess that is something next level. I think CF or bunny will bring that or improve their CDNs. They already can cache some fonts, third party scripts. I am expecting as paid feature that secure loading scripts only from trusted sources like CF or bunny. I am betting not more then 3-6 month to wait.
Yes. But many, if not most of the affected sites where 100% serious businesses. On those you often have have to enable JS to make the site usable. Some random ad financed site are different stories, there you might just move on.
> 100% serious businesses. On those you often have have to enable JS to make the site usable.
It's not all or nothing. NoScript [0] lets you run JS (and other specific types of content) from domains and/or sub-domains you allow and reject from those you don't. So you'd allow JS from my-serious-business domain and their CDN, and disallow all else, including those mentioned in the linked article.
> Some customers serve their snippet from their own domain, such as opt.scheels.com/js/<id>.js or www.vodafone.com.au/js/<id>.js. Those were affected too, because the content comes from the same Optimizely project.
The article states the opposite, which I’m almost certain was due to the attacker realizing that Content-Security-Policy was blocking the payload on many sites:
“A full inline copy of the payload, which runs through eval if 1688360.cc is blocked.”
Even if they weren't (I have not studied the details), when using NoScript and something does not work you need to enable additional domains. It is infeasible to find the minimal necessary set, so you'll end up with a long allow list.
And an allow list where I’ve made decisions about individual exceptions is a problem why? Do you think having a long list of exceptions is a security problem compared to the entire Internet?
Don't know about noscript specifically, but uMatrix has a default block-list that's safe to enable by default and that includes optimizely. So allowing everything except this default block-list would have protected you from this attack already without any inconvenience of managing allow-lists.
This is simply incorrect. I still use umatrix. On bloated sites there's typically only one or two key domains that need to be permitted. Once you've been running it for a while you can usually guess which ones are needed based on what the grid of resources looks like. Most third party domains aren't required most of the time even for what you described above as "serious businesses". I block most of the third party resources on my bank's website. Honestly the number of dependencies that they pull in absolutely disgusts me; their online banking interface absolutely chugs on low end devices.
I ran uMatrix as well for a long time, and in general there were very few sites where you had to enable JS across the board.
Was a pain though on eCommerce sites as they'd often switch payment processors and then you had to do the whole dance of partially successful payments until you found the right set of allow lists.
Agree, it takes about a week to allow like 60% of pages we visit daily. Then over a month the rest of sites that we occasionally visit. Then there are the pages that completely break when we remove the ad js crap, and then you can take a decision on if you want to continue using that site or not. I have also messages a couple of smaller sites that break when you removed googles tracking, because of the order they load JS, and they promptly fixed it.
I often do so regularly, and have built up passive knowledge about what shit to enable and what not. Optimizely is very far in the realms of shit not getting enabled. I have not had a single instance of it being useful for me as a visitor or user of a website. It does not serve me and it's trash.
If by serious businesses you mean ones that view the inclusion of megabytes of unvetted JS as a security risk and a failure of engineering, there are few serious businesses.
I don't disagree. The majority are seriously shitty businesses. Can a name a big bank, hotel reservation, airline, you name it that is significantly different in that respect?
If you are often updating your scripts, you need to offer an SDK so that users can import the code through an SRI checked script. It looks like Optimizely has one available: https://github.com/optimizely/javascript-sdk, so there's no reason not to use SRI with it.
I still don't see how this would prevent this type of attack.
An SDK doesn't necessarily mean that scripts are kept up to date as a site's build could've been done months or years ago with the SDK version pinned in npm.
If I were to setup a system today, I would mirror any script from a third-party and pipe it through a LLM to check for malicious code before deploying.
> We don't know how the attacker got write access to so many unrelated customer projects: a compromised Optimizely system, a compromised credential with cross-customer access, or something else. That's for Optimizely to explain. To be clear, Optimizely's CDN served exactly what it was given; this was not a CDN cache-poisoning attack. Optimizely's own sites were affected as well: the snippets for www.optimizely.com and login.optimizely.com both carried the code.
Interesting that it is not necessarily a supply chain attack a la LiteLLM/some of the other recent breaches. Possibly just a mass compromise of customer passwords.
This stuff is so avoidable but unless we implement meaningful reprocussions things aren't going to change. There is no accountability.
In this case optimizely should have had better systems to spot a release like this (internal cool downs, check and balances, phased rolouts, automatic security checks). And sites that have massive user bases (cnn, bbc, etc..) have zero excuse to not have the same.
I would go so far as to say those sites should not be serving from vendors but should be vendoring their own scripts that they also are passing through security audits and serving them selves with cool downs.
How do these companies not have better dedication to protecting their users is beyond me.
Optimizely is slowly dying ever since they forced new customers into their DXP/Saas solution.
Optimizely, or Episerver as it was previously called, was such a pleasure to work on - but since version 13 they more or less vendor locked you into their own authentication (Opti ID / Entra AD (?)) and their Optimizely Graph for search and content look up. You can keep self-hosting but you won't get access to Opti Id or Graph if you're not on their tenant.
I guess this is the inevitable happen - everyone gets hacked at the same time when hosting on a SaaS platform.
Completely true about their auth, each time I have to use their service, it's incredibly complicated to even login, their policies for password expiration are ridiculous and you are constantly prompted to move your credentials to newer systems without any benefit.
These phrases are not awkward, it's high-quality human English writing.
The models are using language correctly, but their whole universe is limited to a single session. They're not aware that all the previous sessions also used the same phrase, that you've seen it 10 times in the last hour, etc.
Humans naturally avoid repeating the exact same phrases and jokes they just heard from several other sources, but sometimes you can see the same phenomenon with people too: when there's new bullshit phrase or trend spreading through tiktok/instagram (or popular press), there's a brief moment you can catch many normies regurgitating the exact same phrase "independently".
The front of the meme virality wave sure does look weird.
> These phrases are not awkward, it's high-quality human English writing.
It’s not high-quality, though — it’s SEO speak to pad blog posts without adding anything of substance. We notice it more because that stuff is so common in the training data but not the experience of anyone who doesn’t spend time on content mills.
Because if you ignore the surrounding context and any broader technical details that are relevant then the phrasing itself in a vacuum isn't awkward at all. In fact I'd say it reads like well polished marketing copy. The trouble is that not only does it not exist in a vacuum but also all of us are sick of what has now become utterly cliche voicing.
1. The former seems to have a tiny fraction as much detail as the latter.
2. The former seems super more clear to me to be AI generated prose.
3. The latter explicitly continue talks about the actions they took, that they reached out to the affected parties, etc. From a quick scan, it does not appear that the former claims to have done anything themself?
To be honest I'm surprised that only $7040 was stolen in 9 hours. That seems like a low amount for the amount of traffic the affected sites get in those 9 hours.
it would need a pretty unlikely combination; a user would have to visit (say) bbc.com after copying a crypto address, and before they paste it into something else to send funds
That would be low hanging fruit. You can bet they saved all this data and can spend the next months mining it to see which passwords fit on which sites and so on.
I hated using optimizely. Built for the dumbest devs, integrates at the core, optimizely has already burned bridges with their pricing and support, and now this. Sadly, their ignorant users get to suffer consequences of their ignorant SaaS provider
Feels nice knowing several years ago I worked on forcing CSP (including removing unsafe-eval) on some customers for a site builder product, as they have to manually whitelist domains which limits the impact of these sorts of attacks...
Yeah, honestly I wish they'd move that section higher up - it's probably he most genuinely useful tidbit in the article.
As an aside - I'm rabidly against "analytics" companies that come in and suggest that the right "getting started" choice is to drop untrusted JS they own (and can update at any time) onto your site.
It's "easy", but it completely breaks your security model.
If I were a lawyer... I'd be taking a serious look at putting optimizely into the meat grinder for recommending 'unsafe-eval' as a script source... it literally has "unsafe" in the fucking name for a reason. This alone should end the company. It won't, because this space is absolutely FILLED with scummy behavior, but it should.
---
A strict CSP is one of the most meaningful ways you have right now as a web developer to reduce attack surface. People should take advantage of it.
I think the kinds of people with "security models" are not the kinds of people with third-party analytics. They're marketing to people who want it to Just Work.
There's no reason to use SRI any more since caches became partitioned. It won't share the cache between different websites for privacy reasons so you may as well just host the file yourself, and then there's not much reason to use SRI.
SRI helps you make sure the script you're loading is what you initially thought, but doesn't ensure that script doesn't load other scripts, no? You need CSP for that.
Specifying scripts by hashes is always a solid defense that platforms like this don't change their script after compromise - but in practice the whole point of optimizely and similar tools it that it's expected the script will change.
Common examples like analytics tools happily change their script bundles whenever they feel like it without just tagging a new version and letting you upgrade.
I remember a team at my old place bringing down the entire app with insufficient defense around an Optimizely script that failed. Seems bonkers to take on these dependencies and I think it's only going to get worse.
It's a feature. Changes to prod should always be reviewed and this fundamental incompatibility with insecurely hosted JavaScript is a signal that it is a bad idea.
Unfortunately, in this case the insecurely hosted JavaScript is the feature; that is, people who are not responsible for sites going down or breaches want to be able to add things nilly-willy (which therefore necessitates a way to bypass review).
Yes. You install a CSP that says "my site cannot load any scripts from third parties". And then any script you run, you audit thoroughly and then host on your own website, ensuring it doesn't try to load anything third-party.
Never load third-party JavaScript on your site. Zero exceptions.
In theory yes, in practice no since every SaaS wants to ship 5 times a day and add 15 new dependencies on other SaaS providers each release, and marketing wants to load gobs of 3rd party tracking and analytics code from their favorite SaaS of the day without having to get approval from those pesky security people. Let's not even mention ad networks.
Doesn't GTM detect malware, or was this malware only being served to non Google IP addresses? Even so, I'm sure Google has access to millions of residential IP addresses to run security scans.
A good reminder to never allow optimizely's shit to run. Here is to uBlock origin and noscript/uMatrix saving the day. Like probably every day the last 15 years or so.
I visited at least 2 of the listed sites. I don't own any crypto, so no problem there. I might have entered my credit card details to some site, definitely not to the 2 listed ones. But with the whole "load 300 scripts" cancer in "modern" web you never know.
I guess all I can do now is emptying my browser cache to be on the safe side. The writeup (piece of marketing for their product) has no section for affected end users
Yes, I sometimes used NoScript and uBlock Matrix, but gave up on those because it's a tedious job to manage your allow lists.
Perhaps at last these unserious names will be abandoned as newsworthy failures become more common in the industry. Viz., gravitas may start to have obvious market value again.
Well, yes. And that means that the exploit can drop some bit of persistent malware, that will collect things like crypto wallets and passwords, throughout the OS, and not just one website.
Every time one of these new attacks drops, you know there's a couple devs at OpenAI that look at each other and go, "This definitely wasn't us, right? ...right?" And how confident are they that out of Outlook, FAA, banks, and now Optimizely skimmer that none of them were either 1) agents up to some fuckery or 2) someone using their system to get up to some fuckery? It can't be 100% confident, right? ...right?
It's one of those sentences that seems like it's meaningful on first glance but that in reality conveys no information. For whatever reason LLMs are experts at cranking these sentences out.
These are not static scripts. Their content changes frequently. The customer is using them to inject dynamic content into their own sites. What you are describing would not work for this use case.
It's also what they want. They want to change the code frequently, that's the whole point of outsourcing it. Telling people "just don't do the thing you want to do" is not an effective recipe for change.
Yes, AI is making it much easier to hack - you can just spin up a bunch of instances to look for vulnerabilities, you don't even need to do much manual work
Honestly this is getting hilarious. All the shitty decisions everyone made are now coming back to haunt them. It's "I told you so" on a worldwide cyber level.
Before the AI era you could come with a technical/security objection which was correct in theory, but everyone knew the chances of it being exploited were minuscule. In most cases what was to be gained outweighed the costs (this hack collected a mere $7k). Today, saying "this is technically wrong" likely means "it will be used against us".
tl;dr: Someone gained write access to the script Optimizely serves on its customers' websites, and they injected malware.
What is interesting is that the write access wasn't just for a single customer, but hundreds. So either tenancy isn't strong enough, or the level of access involved in the compromise was pretty severe.
the base JS snippet is often the same for all tenants. The script behaviour changes depending the parameters the tenant set on their websites. All the hackers need to do is to gain access to the CDN service where the base snippet is serve, add their payload in the base snippet and it will affect all tenants.
You often don't even need to instruct it to not use 3rd part libraries because in my experience it defaults to creating its own hot mess, furthering your dependence on it
Just stop using JS... If you control the server why not have the vendor have a server level integtration that way you are not serving any remote scripts. It would be interesting to do a study on how many of these compromise emnate from third party scripts and I am willing to bet it is over 90%
The whole concept of Google Tag Manager was that you could give marketing departments the ability to add "tags" to the website without having to involve engineering. That's how we got 10MB of javascript with every news article.
I hope folks are re-evaluating that now.
Client side A/B testing through services like Optimizely and VWO and Adobe Target is a massive security nightmare in of itself. You're giving anyone with access to a third party service the ability to run any code they like on your website, and change any and all interface elements as they see fit.
Even one of these accounts being compromised would be a massive security incident that would potentially put millions of people's data at risk.
Just pointing out that these testing solutions can also be deployed server-side instead of via client-side tags. I suspect that is the route a lot of their customers will take now.
And even at the edge via things like edge side includes or their modern worker versions.
It also makes adding csp a nightmare. Your marketing department is putting in 30 trackers, changing monthly. And then there’s different countries for end clients with different TLDs for connect-src. You can’t possibly get anything locked down enough. That’s bad for security when you rely on csp for mitigating other factors.
> And then there’s different countries for end clients with different TLDs for connect-src.
Oh how I hate that one. I wish Facebook, Instagram, Google and others would figure out serving the public scripts from .com only.
I'm sure they know how to serve from a single domain - but it is in their interest if you are forced to give them the nonce pass. So yeah, it's (again) all about incentives.
relying on any client-side "security" is probably a bad idea
CSP is there to protect your user’s browser, not your servers. It’s ultimately their credentials that are going to get compromised with an XSS attack.
All things considered I think CSP is a good thing, a bit of a nuance for website operators, but it probably helps protect against lot of attack types.
Client side security isn't the issue, but CSP is kind of a shit show. The configuration apparatus is godawful. I think the best way to think of it is if someone turned CSS (yes, the styling language) into a security configuration protocol. It uses the same style of cascading overrides that CSS employs.
What is the alternative when you need to embed third-party JavaScript on your pages? Other than just "not doing that"?
If it's available, pull the third party JS in via package manager rather than a CDN you don't control.
NPM isn't perfect but there's typically a substantial delay between the release of a malicious package and that code getting onto your production environment. Unless you're automatically updating every day, there's a good chance someone will discover the problem before it affects you. And you can always scan your upstream dependencies.
This doesn't work if you're using someone else's CDN. Any change to their content is distributed to all your users immediately.
A certificate-based approach for scripts residing on different domains, for example, so you can be sure it hasn't been tampered with.
A javascript file served from example-a.com should be able to run on example-b.org only if it has been signed with the same TLS key used for example-a.com .
CSP allows you to configure your loaded JS with a complementary hash check. If you're the one who puts the script on your page, in theory, it should be safe for as long as remains the same.
Isn't this already true if the script is served over HTTPS?
How would that actually look in reality?
Vet everything, cut down to 2-3 services tops. Selfhost analytics whenever possible. Point to the OP article when asked why we can’t blindly put 30 tags using GTM.
Response to your pointing out: "so what?"
> Response to your pointing out: "so what?"
Then you've done your job and move on. Just make sure that there's enough public record of your warning that - when the next cdn supply chain attack happens - you can say "I warned you about this and you decided not to take action".
No organization is perfect. I don't think this is something to quit over. Just make sure to CYA.
Response: "Here's my two weeks notice for leaving, let me know if you want to start caring about users again".
Life's too short to work for people who don't care about others.
You go ahead and find me another job then.
Life’s too short to care about random strangers on the internet sometimes.
Their response: "ok"
and things continue being insecure
Other than tracking, what situations are there where "not doing that" is not an option?
Not doing that. It is just not a secure thing to do.
Proxy through your own domain? Has its own issues though
That is not any more secure if you are just blindly proxying the third party scripts and in fact is worse since it runs in the context of your own domain.
Since the third party tag can request anything, this doesn't stop anything and now your domain has an open redirect.
Most of it is server to server now with server side tagging. 3rd party cookies are now 1st party and beacon call is server to server.
I mean, really that situation is CSP finally earning its keep. Put whatever you want in tag manager but at the end of the day it's not going to work unless I review it.
Google’s suggested workaround for CSP enabled sites for GTM is to use a nonce, which GTM then propagates to all further scripts that it injects.
https://developers.google.com/tag-platform/security/guides/c...
Nonce would kill this idea, but fortunately it is the webadmin that needs to install it.
In the end this is a tradeoff between security and convenience.
> I hope folks are re-evaluating that now.
Not going to happen. Most decisions in tech do not factor in risk, only cost. Risk mitigation usually requires considerable spend, and very few people/companies are willing to spend on low-probability events.
I know you're sort of speaking abstractly, and I think you're correct that this is how companies interpret risk, but it's interesting to give a moment's thought to -- Murphy's Law tells us all events have a 100% probability of occurring given a certain amount of time. And we tend to see that proven out time and again.
This might change as AI-assisted hacks and outages become more frequent. A materialized risk with associated costs is very different from an abstract risk that's hard to quantify.
There were a couple of years before the concept of tag manager where publishers would just blindly call a JS file on their trafficker's server. This file was maintained by engineering, and you would still end up with MB of junk on every page (especially since this was before any kind of tracking regulation, wild things would happen behind the scenes). So this is more on the advertising industry itself than on wannabe devs PMs.
Beyond all the issues with third party JavaScript inclusion on sites blowing up their security perimeter, there are also significant issues with the link tracking for these as they nest and re-encode their values so much in the URLs any reasonable firewall concludes that they're some kind of obfuscation attack.
We also get wasted Engineering time tracking down bugs caused by crap being added.
We are indeed re-evaluating it now
What is that saying? A person will not understand something that their paycheck requires them to not understand.
That’s why it needs to become criminal for a website to serve malware, including through third party channels like ad networks and analytics services. Serving shit websites is no different than knowingly, or negligently, dumping toxic waste and should be treated the same.
If it's criminal for sites to accidentally serve malware, what does it mean for open source projects to accidentally get hit by supply chain attacks, or for them to get compromised by a rogue contributor (eg. xz)?
Those website operators include third-party code willingly and explicitly. There was never an accident. There is no supply-chain attack here: you cannot seriously claim that unaudited, unauthenticated terminal with root security principal getting used is an attack.
If your website is making any requests outside of the original domain, there's no "accidental" in anything the website does.
Someone _could_ argue that a vibecoded website is doing "accidental" things (like serving malware).
Someone could argue that the sky is green.
Malware is malware. People shouldn’t be offering software that they can’t stand behind.
I see no way to distinguish between huge megacorps and individuals.
A business in the US has legal liability if they fail to do their due diligence in their supply chain and it turns out their suppliers are involved in human trafficking, child labor, forced labor, etc. or is using hazardous materials in the product.
There should be standards that people have to live up to if they are going to provide a product, even for free, to others.
Do they? Without signing up?
Any cases of that being enforced with any significant business ever? What about aiding genocide? (FB, and weapon makers for example)
One could distinguish between criminal negligence and forgivable negligence in court. Unfortunately for GP that probably means you can't actually stick charges on anybody who serves malware via an ad network like this.
> One could distinguish between criminal negligence and forgivable negligence in court.
Open source is already expensive from a time standpoint, you wanna make it expensive from a monetary standpoint too?
Yes. Why should you be allowed to be negligent just because your product is open source?
The question is more "why should it cost $50k in legal fees to prove I wasn't negligent in my unpaid hobby".
in theory at least, frivolous lawsuits should be compensated for
Software is already hard from functionality standpoint, you wanna make it hard from security standpoint too?
It is already hard from a security standpoint, only that we ignore security.
Easy solution: Don't use ad networks. That'll likely free your site from illegal tracking, as well as ad trash, and will even improve your lIgHtHoUsE score!
Yes. It means the legitimate maintainer of xz would be in prison right now.
Liability should aggregate up to the largest platform involved. Optimizely in this case, I guess. Google for Google Tags.
We should also enforce antitrust laws.
> A person will not understand something that their paycheck requires them to not understand.
I never really understood this idiom, because how can a PHB engineering manager (for example) be required to not understand their team's technology in order to retain their job? Surely if that manager knows their team's work inside out then they would be a more effective manager - which is good for everyone?
If I squint-my-brain, I can see how (for example) that anyone who works full-time as an SEO "consultant" in the year 2026 would only be if they actually believed that SEO tricks actually worked, which requires them to not understand how modern search-engines' spiders actually work (because it's my belief that "SEO" has been either nonsense bollocks - or a complete grift - since the mid-2010s). Am I close?
Because it is not the PHB that starts the problem. It is someone from some other department (usually marketing) who wants tracking, wants it now and wants it done with GTM and will whine to th C-levels until the engineers add Google Tag Manager.
It's not the team's technology that's being referred to. It's about hazards. People whose job it is to do X have a hard time understanding that X shouldn't be done.
In this case, it's the dangers of the tracking and advertising frameworks being used for hostile purposes.
> because it's my belief that "SEO" has been either nonsense bollocks - or a complete grift - since the mid-2010s
It's a continuous case of whack-a-mole. SEO consultants think <thing> would be a good way to improve the search ranking of their customers, it works, it spreads throughout the industry, Google catches on, rinse and repeat.
As soon as something is a KPI, there is incentive to game the system - with the top end of "gaming the system" toeing or crossing the line into fraudulent, anti-competitive or otherwise illegal territory.
If you're only still there because your boss believes SEO works, you better keep acting like your SEO tricks work.
It's not about actual understanding, but about pretending not to understand — because if one does, they might be out of a job. ;)
"It is difficult to get a man to understand something, when his salary depends on his not understanding it."
--Upton Sinclair
"There are none so blind as those who will not see" - John Heywood (et al.)
The follow on line from that one is even better: "The most deluded people are those who choose to ignore what they already know". I would probably want to add on things like "because it's not convenient" or similar for particular relevance to some of the stuff I've been dealing with recently at work. Folks know it's dangerous to do something a certain way, and know how to do it a safe way, but it's not as convenient so they delude themselves it'll be fine just this once. And just this once too. And just this once, once more. Slowly but surely normalising the deviance.
In the context of the Optimizely case, people know they're giving it dangerous levels of capabilities, but it's just too darn inconvenient to do it in a safe way so sure, why not. It'll be fine (and it will be, right up until it isn't and you realise your choice has compromised your users)
> it needs to become criminal
I know we really like mass incarceration in America. But maybe start with civil liability and a private right to enforcement?
Even just the process of going to court usually scares people straight even if they don't end up with any liabilities or costs.
this. amen
The funniest part is adding GTM generally means you score low on google's pagespeed metrics through lighthouse
GTM is incredible, really.
I'm programming for over 30 years now and GTM is my personal #1 most hated software.
Really? Worse than IE6? I'd say it's a close call, but IE6 "wins".
Well, GTM is my #1 in living software. MSIE6 is dead. But yeah if this filter is being removed then MSIE6 takes the cake obviously.
Just had marketing try to add all OTHER previous page events to a new event, because they didn't want to to go through the trouble of creating a dashboard using the user funnel to this feature.
They knew they could do it, and how, they just didn't want to spend 10 minutes doing it themselves. To them, asking an engineer to do it in code is easier because it only takes a minute to ask.
But the engineer doing it is doing it once rather than the marketeer doing it manually every time..
The marketer just needs to hit "Save" once.
I am sure. Marketing folks are famous about being reasonable and showing restraint, even if their revenues would suffer.
Marketing folks they were the driving force behind Adobe Flash, but the issues with that were so great that the industry did eventually get rid of it. It just takes time for the costs to become sufficiently apparent to everyone.
AI-assisted hacks and outages are accelerating the process of making the costs of weak security apparent.
The “insights” get product managers promoted, so it’s here to stay.
Unless we get rid of product managers, which isn’t a terrible idea.
Good luck with that.
The owners of the ad platforms end up dictating what data is used, they have the leverage and also marketing departments with power end up forcing engineering to use the audience tools Google and Meta provide. At least GTM gets blocked by ad-blockers.
We got 10MB of javascript (maybe 10 years ago...) because engineering needs to build one frontend with multi browser and multi screen support so they picked an off-the-shelf do-everything framework like Amber or Angular, or a react/redux build with 300 nested dependencies.
No, the 10MB (sometimes way more) of JavaScript for a damn news article has nothing to do with SPA frameworks (Angular) or libraries like React. It is tracking malware put there by marketing/advertising idiots.
I remember when this first became a thing when page size was important. I fought tooth and nail with marketing to try and block tag manager. I've had third party URLs block production apps in Chrome before because some marketing url was hosting malicious code.
The most infuriating thing, when you do remove these tags, marketing forgets they exist and are like "Oh we don't use that anymore".
About the "Oh we don't use that anymore.": That shows you the professionalism of the profession.
Often also true for software development, unfortunately.
Porqué no los dos? Both are true. Heavy JS frameworks, ads, trackers, etc., but also heavy web fonts, of which maybe 5 icons are used for the site, but the font weighs a megabyte. Just in case you need it.
Those were decisions at a time engineering was time constrained. All those ancillary things are returning to engineering since its just one more prompt now to manage it.
This stuff is so avoidable but unless we implement meaningful reprocussions things aren't going to change. There is no accountability.
In this case optimizely should have had better systems to spot a release like this (internal cool downs, check and balances, phased rolouts, automatic security checks). And sites that have massive user bases (cnn, bbc, etc..) have zero excuse to not have the same.
I would go so far as to say those sites should not be serving from vendors but should be vendoring their own scripts that they also are passing through security audits and serving them selves with cool downs.
How do these companies not have better dedication to protecting their users is beyond me.
> How do these companies not have better dedication to protecting their users is beyond me.
It doesn't cost them anything but a "whoops" post when they fuck up, so the incentive is misaligned (as you noted under repercussions).
The cost is externalised onto everyone else not them so it's not going to change.
This is the only answer
Are you sure? The affected companies are going to totally make Optimizely pay for damages. That could become a lot of money pretty fast. And tbh, I'm not sure that the average ToS's maximum liability ceiling will hold up in court with a blast radius this big.
Usually, the plaintiff/claimant has to prove the damages they suffered, not the damages their (non-paying) users suffered. So then they would have to claim reputation damage or something like that, and that might not be so easy to prove.
It could happen but if you look at the last few decades of losses from software bugs, it’s exceedingly rare for the cost to be more than lost future business, and even that’s unusual. Remember when CrowdStrike brought down businesses around the world and they suffered … having their share price more than double?
https://finance.yahoo.com/quote/CRWD/
Accountability is incompatible with tech sector as it currently exists. The entire ethos of move fast and break things is incompatible with it never mind the shady adtech part of the sector.
Accountability is incompatible with maximising next qtr profits. Modern business culture is rotten.
Partially but don’t think that quite follows. A lot of businesses aren’t as fraught with conflicts of interest as tech.
A company making say bricks has an incentive to cut costs too certainly but they’re not implementing loot boxes, tracking people semi covertly, implementing dark patterns to boost engagement by any means or heaven forbid launching missies based on sketchy AI decisions.
> There is no accountability.
When Europe tries they are either laughed at or accused of slowing down innovation.
This wasn’t a release. It was their engine that emits the variants being compromised.
Please tell me in whose incentive it is to change any of this and hold the sites accountable?
Who cares? Next week you forget about this and move on
> I would go so far as to say those sites should not be serving from vendors but should be vendoring their own scripts that they also are passing through security audits and serving them selves with cool downs.
This is the bare minimum. Automatic 3rd party requests should be considered data unsolicited data disclosure under GDPR.
None of that is part of commercial web dev culture. Leftpad-level dependencies are.
Maybe I overlooked it, but is this something persistent that I should take care of on my phone and computer or was it confined to a browser session?
And would uBlock origin have prevented it?
There's a missing step in the explanation. Do major websites like CNN and BBC handle cryptocurrency or credit cards on the same website? Why? Or is there also a browser security bug?
I had the same question, and I really wish these reports would make in clearer what exactly the malware can and can't do. The whole point of web browsers is that sites run in a sandbox so that even if a site does serve malicious code, it should be limited to just that site.
From what I could tell, the malware just generally looked at anything on the page, in cookies/local storage, or clipboard that looks like a crypto address or credit card number. So for sites like CNN or BBC they shouldn't be able to do much damage unless there is a subscription page for example that takes credit card number.
However, the malware also delivered the Coruna and DarkSword iOS exploit kits, which do take advantage of browser vulnerabilities in iOS to cause more serious damage.
The report is appalling. The scale of this is enormous. Saying to update your stuff is not helping detect infection.
I feel that whole web site model is broken. They are run in "sandbox" same way AI companies run their sandboxes. Not caring at all where they do call... So every other site needs to implement protections against other sites.
The model is broken on fundamental level and it will never be fixed.
the toolkit appears to also skim credit cards and there are many e-commerce pages in that list.
I imagine CNN does, but they might have been smart enough to unburden the payment pages from the marketing/surveillance cruft. Or perhaps not. There have also been browser escape vulnerabilities.
I've never understood why Optimizely is a thing. Why is A/B testing so hard? I implemented a full A/B test system for a major media website in 1 day about 10 years ago and its still in use today (I can pull the the secret report on any page and I just did). Its just simple induction algorithms, tagging and conditional rendering.
Marketing teams don't have engineering resources, generally speaking. But they do have the clout to get engineering to install this little tool that will help them "self service."
NewsWeek has listed 150+ popular websites across major economy sectors.
https://www.newsweek.com/cyber-attack-websites-affected-1272...
This is likely a result of Optimizely cutting headcount on their legacy product engineering teams to reallocate funding towards their AI product. We’re going to see more stories like this for every company taking a similar approach.
I guess that is something next level. I think CF or bunny will bring that or improve their CDNs. They already can cache some fonts, third party scripts. I am expecting as paid feature that secure loading scripts only from trusted sources like CF or bunny. I am betting not more then 3-6 month to wait.
All sites detected to be using Optimizely: bloomberry.com/data/optimizely-web-experimentation/
This is exactly the kind of use case for NoScript - reduce exposure to potentially malicious content.
Yes. But many, if not most of the affected sites where 100% serious businesses. On those you often have have to enable JS to make the site usable. Some random ad financed site are different stories, there you might just move on.
> 100% serious businesses. On those you often have have to enable JS to make the site usable.
It's not all or nothing. NoScript [0] lets you run JS (and other specific types of content) from domains and/or sub-domains you allow and reject from those you don't. So you'd allow JS from my-serious-business domain and their CDN, and disallow all else, including those mentioned in the linked article.
[0] https://noscript.net/
Sure I know, I tried NoScript in the past.
It would not have helped here. gitlab.com and my bank are on the list. Obviously I could not have blocked them.
> gitlab.com and my bank are on the list.
Per the linked article, those are not the domains where the malicious JS was served from.
> Some customers serve their snippet from their own domain, such as opt.scheels.com/js/<id>.js or www.vodafone.com.au/js/<id>.js. Those were affected too, because the content comes from the same Optimizely project.
in some cases they were
The article states the opposite, which I’m almost certain was due to the attacker realizing that Content-Security-Policy was blocking the payload on many sites:
“A full inline copy of the payload, which runs through eval if 1688360.cc is blocked.”
Even if they weren't (I have not studied the details), when using NoScript and something does not work you need to enable additional domains. It is infeasible to find the minimal necessary set, so you'll end up with a long allow list.
And an allow list where I’ve made decisions about individual exceptions is a problem why? Do you think having a long list of exceptions is a security problem compared to the entire Internet?
Not talking about a security problem. It's a workload problem for me.
Don't know about noscript specifically, but uMatrix has a default block-list that's safe to enable by default and that includes optimizely. So allowing everything except this default block-list would have protected you from this attack already without any inconvenience of managing allow-lists.
This is simply incorrect. I still use umatrix. On bloated sites there's typically only one or two key domains that need to be permitted. Once you've been running it for a while you can usually guess which ones are needed based on what the grid of resources looks like. Most third party domains aren't required most of the time even for what you described above as "serious businesses". I block most of the third party resources on my bank's website. Honestly the number of dependencies that they pull in absolutely disgusts me; their online banking interface absolutely chugs on low end devices.
I ran uMatrix as well for a long time, and in general there were very few sites where you had to enable JS across the board.
Was a pain though on eCommerce sites as they'd often switch payment processors and then you had to do the whole dance of partially successful payments until you found the right set of allow lists.
Agree, it takes about a week to allow like 60% of pages we visit daily. Then over a month the rest of sites that we occasionally visit. Then there are the pages that completely break when we remove the ad js crap, and then you can take a decision on if you want to continue using that site or not. I have also messages a couple of smaller sites that break when you removed googles tracking, because of the order they load JS, and they promptly fixed it.
I often do so regularly, and have built up passive knowledge about what shit to enable and what not. Optimizely is very far in the realms of shit not getting enabled. I have not had a single instance of it being useful for me as a visitor or user of a website. It does not serve me and it's trash.
However many will put optimizely and things like that proxying through their domains.
If by serious businesses you mean ones that view the inclusion of megabytes of unvetted JS as a security risk and a failure of engineering, there are few serious businesses.
I don't disagree. The majority are seriously shitty businesses. Can a name a big bank, hotel reservation, airline, you name it that is significantly different in that respect?
Yes, but unfortunately I allow-listed optimizely some time ago because some sites wouldn't work without it.
I think uMatrix Origin blocks optimizely by default. It's more user friendly than NoScript.
Subresource Integrity has been available for over a decade. It’s time to start implementing it with these services.
https://developer.mozilla.org/en-US/docs/Web/Security/Defens...
I think it's not used because they are continuously updating their scripts for users.
If you are often updating your scripts, you need to offer an SDK so that users can import the code through an SRI checked script. It looks like Optimizely has one available: https://github.com/optimizely/javascript-sdk, so there's no reason not to use SRI with it.
I still don't see how this would prevent this type of attack.
An SDK doesn't necessarily mean that scripts are kept up to date as a site's build could've been done months or years ago with the SDK version pinned in npm.
If I were to setup a system today, I would mirror any script from a third-party and pipe it through a LLM to check for malicious code before deploying.
> We don't know how the attacker got write access to so many unrelated customer projects: a compromised Optimizely system, a compromised credential with cross-customer access, or something else. That's for Optimizely to explain. To be clear, Optimizely's CDN served exactly what it was given; this was not a CDN cache-poisoning attack. Optimizely's own sites were affected as well: the snippets for www.optimizely.com and login.optimizely.com both carried the code.
Interesting that it is not necessarily a supply chain attack a la LiteLLM/some of the other recent breaches. Possibly just a mass compromise of customer passwords.
Optimizely is a pretty garbage platform so this is unsurprising. Love the work though folks like report uri are doing finding this.
This stuff is so avoidable but unless we implement meaningful reprocussions things aren't going to change. There is no accountability. In this case optimizely should have had better systems to spot a release like this (internal cool downs, check and balances, phased rolouts, automatic security checks). And sites that have massive user bases (cnn, bbc, etc..) have zero excuse to not have the same.
I would go so far as to say those sites should not be serving from vendors but should be vendoring their own scripts that they also are passing through security audits and serving them selves with cool downs.
How do these companies not have better dedication to protecting their users is beyond me.
> On iPhone/iPad/iPod only: injects boot.js from two .icu hosts and, on iOS ≤ 18.6, a hidden iframe.
Anyone knows what this might be? Can you jailbreak from safari on those versions?
[edit] From the other link, it seems to be Coruna and DarkSword exploits
Why should browsers be allowed to read or write user files without user intervention and serious warnings?
Here the attackers targeted forms, swapping crypto addresses or exfiltrating CC details, they did not read or write user files.
Optimizely Status site: https://status.optimizely.com/
Currently says "Partial Outage Affecting Recommendations Tracking and Reporting"
It’s not that issue, it’s this earlier one: https://status.optimizely.com/incidents/ftw0fh9wx43z
Optimizely is slowly dying ever since they forced new customers into their DXP/Saas solution.
Optimizely, or Episerver as it was previously called, was such a pleasure to work on - but since version 13 they more or less vendor locked you into their own authentication (Opti ID / Entra AD (?)) and their Optimizely Graph for search and content look up. You can keep self-hosting but you won't get access to Opti Id or Graph if you're not on their tenant.
I guess this is the inevitable happen - everyone gets hacked at the same time when hosting on a SaaS platform.
Completely true about their auth, each time I have to use their service, it's incredibly complicated to even login, their policies for password expiration are ridiculous and you are constantly prompted to move your credentials to newer systems without any benefit.
All of that for them to mess their own security.
If it was on me, I would stop using them.
Unfortunate with this nonsense slop report for a seemingly significant event like this.
> Your origin can be clean while your visitors are robbed — the compromise sits in the tag layer.
Why do LLMs still produce such unusually awkward phrases, despite being trained on human English?
Model collapse
These phrases are not awkward, it's high-quality human English writing.
The models are using language correctly, but their whole universe is limited to a single session. They're not aware that all the previous sessions also used the same phrase, that you've seen it 10 times in the last hour, etc.
Humans naturally avoid repeating the exact same phrases and jokes they just heard from several other sources, but sometimes you can see the same phenomenon with people too: when there's new bullshit phrase or trend spreading through tiktok/instagram (or popular press), there's a brief moment you can catch many normies regurgitating the exact same phrase "independently".
The front of the meme virality wave sure does look weird.
> These phrases are not awkward, it's high-quality human English writing.
It’s not high-quality, though — it’s SEO speak to pad blog posts without adding anything of substance. We notice it more because that stuff is so common in the training data but not the experience of anyone who doesn’t spend time on content mills.
But it's not especially or high quality English either. That phrasing is awkward even in isolation.
No it is not, imo. It is dense, terse, and complete. It is in fact high quality geek speak.
Agree to disagree! Fair enough.
It's not suitable for a technical writing, it's too engaging for what it has to say.
>it's high-quality human English writing
AI truly appeals to the most brain dead, ignorant mouth breathers, textbook evidence #666
Because if you ignore the surrounding context and any broader technical details that are relevant then the phrasing itself in a vacuum isn't awkward at all. In fact I'd say it reads like well polished marketing copy. The trouble is that not only does it not exist in a vacuum but also all of us are sick of what has now become utterly cliche voicing.
there's no such thing as a "tag layer"
That sounds like a technical detail to me.
If my Origin was "clean" then it wouldn't be loading any scripts from external websites...
ahhh yes, the famous tag layer
That's layer 8 of the OSI model.
It’s right above the egg layer.
Still better than the human slop filter Gizmodo ran it through in their rehashed article. Actual quote:
> On mobile, Report URI says they attempted to install exploit kids.
dupe 2 hours ago, 15 comments: https://news.ycombinator.com/item?id=50523300
it's been merged
Nope
Hm, which one is the original? The merged article actually looked very informative: https://amirulcyber.com/2026/11/16/optimjack-optimizely-skim...
whereas the one that we are commenting on now looks like an ad: https://blog.report-uri.com/worlds-largest-brands-infected-w...
It opens with:
> This is exactly the kind of attack Report URI is designed to detect. Our platform can ...
It's otherwise nearly the same information.
It would be kind of ironic if the hijacking warning article got hijacked, I guess?
1. The former seems to have a tiny fraction as much detail as the latter.
2. The former seems super more clear to me to be AI generated prose.
3. The latter explicitly continue talks about the actions they took, that they reached out to the affected parties, etc. From a quick scan, it does not appear that the former claims to have done anything themself?
To be honest I'm surprised that only $7040 was stolen in 9 hours. That seems like a low amount for the amount of traffic the affected sites get in those 9 hours.
it would need a pretty unlikely combination; a user would have to visit (say) bbc.com after copying a crypto address, and before they paste it into something else to send funds
That would be low hanging fruit. You can bet they saved all this data and can spend the next months mining it to see which passwords fit on which sites and so on.
Where is the phone home endpoint 1688360.cc hosted? Registar?Doesn’t seem to be any DNS records on that domain (probably updated).
> "and on iPhones and iPads it [the malware] delivered the Coruna and DarkSword iOS exploit kits."
Isn't this anomalous? using RCE root chains in mass-delivered malware? Seems like a new precedent.
There was a report on that last week: https://cyberscoop.com/spyware-attacks-trend-everyday-people...
I hated using optimizely. Built for the dumbest devs, integrates at the core, optimizely has already burned bridges with their pricing and support, and now this. Sadly, their ignorant users get to suffer consequences of their ignorant SaaS provider
Is there anything useful served from any *.icu domain? I wonder if I just should block the whole TLD and be done with it.
I think blocking all but ~6-7 tlds should be pretty safe. I have personally seen so rarely malware or phishing hosted in famous tlds.
Risk is when *.icu is guaranteed to be ICU.
Sorry I had to say it ;-)
Let’s say I use Optimizely (I don’t), is there a good best practice to say?:
“yeah load this JS m, but it can’t load / random other JS referencing sketchy places”
Besides specifying scripts by hashes, you can also whitelist various domains: https://developer.mozilla.org/en-US/docs/Web/HTTP/Guides/CSP (and https://developer.mozilla.org/en-US/docs/Web/HTTP/Reference/...)
Feels nice knowing several years ago I worked on forcing CSP (including removing unsafe-eval) on some customers for a site builder product, as they have to manually whitelist domains which limits the impact of these sorts of attacks...
This writeup https://blog.report-uri.com/worlds-largest-brands-infected-w... has a section on how CSP helped (or didn't) in this attack.
Yeah, honestly I wish they'd move that section higher up - it's probably he most genuinely useful tidbit in the article.
As an aside - I'm rabidly against "analytics" companies that come in and suggest that the right "getting started" choice is to drop untrusted JS they own (and can update at any time) onto your site.
It's "easy", but it completely breaks your security model.
If I were a lawyer... I'd be taking a serious look at putting optimizely into the meat grinder for recommending 'unsafe-eval' as a script source... it literally has "unsafe" in the fucking name for a reason. This alone should end the company. It won't, because this space is absolutely FILLED with scummy behavior, but it should.
---
A strict CSP is one of the most meaningful ways you have right now as a web developer to reduce attack surface. People should take advantage of it.
There's a reason browser vendors now force basically all extensions into a strict CSP by default... https://developer.chrome.com/docs/extensions/reference/manif... They won't even let you manually relax it anymore.
I think the kinds of people with "security models" are not the kinds of people with third-party analytics. They're marketing to people who want it to Just Work.
Optimizely is widely used within many large Saas companies. Including those with literal legal security requirements (soc 2, iso 27001, etc).
So are equivalent companies that recommend similar strategies. Sprig, segment, amplitude, launchdarkly, heap, full story, etc...
It's a very long list.
They all operate on the same strategy. Pitch to execs that they solve real problems and they have basically no eng cost to implement.
All you need to do is inject them into your page with absolutely no restrictions and let them run any scripts they'd like.
Yes: https://developer.mozilla.org/en-US/docs/Web/Security/Defens...
There's no reason to use SRI any more since caches became partitioned. It won't share the cache between different websites for privacy reasons so you may as well just host the file yourself, and then there's not much reason to use SRI.
SRI helps you make sure the script you're loading is what you initially thought, but doesn't ensure that script doesn't load other scripts, no? You need CSP for that.
Specifying scripts by hashes is always a solid defense that platforms like this don't change their script after compromise - but in practice the whole point of optimizely and similar tools it that it's expected the script will change.
Common examples like analytics tools happily change their script bundles whenever they feel like it without just tagging a new version and letting you upgrade.
I wish SRI could be used more,
I remember a team at my old place bringing down the entire app with insufficient defense around an Optimizely script that failed. Seems bonkers to take on these dependencies and I think it's only going to get worse.
It's a feature. Changes to prod should always be reviewed and this fundamental incompatibility with insecurely hosted JavaScript is a signal that it is a bad idea.
Unfortunately, in this case the insecurely hosted JavaScript is the feature; that is, people who are not responsible for sites going down or breaches want to be able to add things nilly-willy (which therefore necessitates a way to bypass review).
this will prevent an unexpected version but doesn't stop a legit but compromised include from loading nefarious resources
Very cool thank you.
Yes. You install a CSP that says "my site cannot load any scripts from third parties". And then any script you run, you audit thoroughly and then host on your own website, ensuring it doesn't try to load anything third-party.
Never load third-party JavaScript on your site. Zero exceptions.
In theory yes, in practice no since every SaaS wants to ship 5 times a day and add 15 new dependencies on other SaaS providers each release, and marketing wants to load gobs of 3rd party tracking and analytics code from their favorite SaaS of the day without having to get approval from those pesky security people. Let's not even mention ad networks.
> This is exactly the kind of attack Report URI is designed to detect.
So is this article just an ad? It makes it hard to tell what's true and what's exaggerated.
The dangers of Javascript
Doesn't GTM detect malware, or was this malware only being served to non Google IP addresses? Even so, I'm sure Google has access to millions of residential IP addresses to run security scans.
what are the practical effects of this? Does this mean that a customer of Zoom was downloading malware all day?
A good reminder to never allow optimizely's shit to run. Here is to uBlock origin and noscript/uMatrix saving the day. Like probably every day the last 15 years or so.
I visited at least 2 of the listed sites. I don't own any crypto, so no problem there. I might have entered my credit card details to some site, definitely not to the 2 listed ones. But with the whole "load 300 scripts" cancer in "modern" web you never know.
I guess all I can do now is emptying my browser cache to be on the safe side. The writeup (piece of marketing for their product) has no section for affected end users
Yes, I sometimes used NoScript and uBlock Matrix, but gave up on those because it's a tedious job to manage your allow lists.
I still think that running scripts in a browser is a bad idea, but what do I know.
I don't know what you know, but I know something you don't know: how to maximize revenue.
14:23 UTC, still no mention of this on their home page.
https://www.optimizely.com/
Perhaps at last these unserious names will be abandoned as newsworthy failures become more common in the industry. Viz., gravitas may start to have obvious market value again.
Compromised? They ate grilled cheese off the radiator or what?
"software is solved" :brainlet:
I mean, yes to all the things, CSP, malware checker, CDN etc.
But how do you know that you personally didn't get affected? Check here: https://karlnorling.github.io/optimizely-hack-checker/
I'm confused. What exactly would be the point of the iOS exploits? Wouldn't the device just become jailbroken?
Well, yes. And that means that the exploit can drop some bit of persistent malware, that will collect things like crypto wallets and passwords, throughout the OS, and not just one website.
Optimizely is such crap... like, all it does is A/B test different landing pages/etc for "conversions." We don't need this.
Every time one of these new attacks drops, you know there's a couple devs at OpenAI that look at each other and go, "This definitely wasn't us, right? ...right?" And how confident are they that out of Outlook, FAA, banks, and now Optimizely skimmer that none of them were either 1) agents up to some fuckery or 2) someone using their system to get up to some fuckery? It can't be 100% confident, right? ...right?
Aren't these all mathematicians who have never had to toil in the operations trenches fixing this stuff, thats why they make 4x my salary right
Can't wait until they start killing hardware
> The malware was a multi-purpose kit.
Somehow this sentence immediately flags this article as AI slop.
It's one of those sentences that seems like it's meaningful on first glance but that in reality conveys no information. For whatever reason LLMs are experts at cranking these sentences out.
Next up the list I'm sure: Sentry.
This has been in my HOSTS file for a while:
...and now I have another reason to keep it there.Reads like an ad to me.
Reminder that this problem has been solved for a while now if admins were not so lazy: https://developer.mozilla.org/en-US/docs/Web/Security/Defens...
These are not static scripts. Their content changes frequently. The customer is using them to inject dynamic content into their own sites. What you are describing would not work for this use case.
Loading dynamically changing code from a third party server is a recipe for disaster, which is exactly what was cooked here.
The code should be static and the configuration (data-only) the only thing that changes frequently, but that is less flexible and harder to implement.
It's also what they want. They want to change the code frequently, that's the whole point of outsourcing it. Telling people "just don't do the thing you want to do" is not an effective recipe for change.
Reading this thread has made me realize that intelligence alone is no garlic, cross, or silver bullet against the bored and conspiracy minded.
Optimizely is ~~infected with~~ malware FTFY!
Is it just me or is there an increase of these events or increase of these posts?
Yes, AI is making it much easier to hack - you can just spin up a bunch of instances to look for vulnerabilities, you don't even need to do much manual work
Some of it's real, the rest is a psy-op. It's not hard to figure out what the goal is.
Sorry, what is the goal?
Honestly this is getting hilarious. All the shitty decisions everyone made are now coming back to haunt them. It's "I told you so" on a worldwide cyber level.
Before the AI era you could come with a technical/security objection which was correct in theory, but everyone knew the chances of it being exploited were minuscule. In most cases what was to be gained outweighed the costs (this hack collected a mere $7k). Today, saying "this is technically wrong" likely means "it will be used against us".
tl;dr: Someone gained write access to the script Optimizely serves on its customers' websites, and they injected malware.
What is interesting is that the write access wasn't just for a single customer, but hundreds. So either tenancy isn't strong enough, or the level of access involved in the compromise was pretty severe.
the base JS snippet is often the same for all tenants. The script behaviour changes depending the parameters the tenant set on their websites. All the hackers need to do is to gain access to the CDN service where the base snippet is serve, add their payload in the base snippet and it will affect all tenants.
In the age of AI if you use third party libraries you deserve all hell unleashed on you
First thing I add to my agents.md is do not use third party libraries, frameworks, fonts, icons, etc unless strictly necessary, ask first
You often don't even need to instruct it to not use 3rd part libraries because in my experience it defaults to creating its own hot mess, furthering your dependence on it
This was inevitable and not entirely novel.
https://x.com/tehlike/status/1432082161125888003 https://x.com/tehlike/status/1467033965215109122
@tehlike is protected
Only followers this account has approved can see its posts.
Sorry my bad. Opened. Not a particularly novel insight either.
Musk-free links:
https://xxcancel.com/tehlike/status/1432082161125888003
https://xxcancel.com/tehlike/status/1467033965215109122
Just stop using JS... If you control the server why not have the vendor have a server level integtration that way you are not serving any remote scripts. It would be interesting to do a study on how many of these compromise emnate from third party scripts and I am willing to bet it is over 90%
stop using scripts fullstop the web was never intended to send program code
any program code should run on the webserver
browsers should be minimalist and only support html
and XUL