On a related note they opensourced an email client: https://github.com/cloudflare/agentic-inbox

That's not a well kept secret, that's just a workflow that almost nobody would accept for their email setup which is the center of most people's digital identify and should always work and not be a duct taped construct to save a couple of bucks.

I assume the constructive use case is some non-techy person asking ChatGPT.

> Hey, please make me a website about my dog woofy. Give it the link myfluffywoofy.dog ;) Thank you!

cf bought an email security company a couple years ago so wouldn’t shock me they have good spam filtering.

I think this post needs to be put in context, for months now Cloudflare has been releasing products that allow their whole platform to be usable by agents with the main objective of enabling their customers to dynamically write code using Cloudflare, this is just another step.

For example, you can now with Artifacts and Dynamic Workers make a lovable-style SaaS where your customers ask the AI agent to write software for them, the agent can run it in sandboxes with no build step, it can version it with a git-compatible API, and now you can even have it buy a domain for the end customer or set up their own cloudflare account when they want to move to production.

I personally have no use case for creating domains via agents, but some of the other features they're releasing around this area are extremely useful and I've started to ship internal tools for my clients where they are used, like giving them their own mini claude code that only does one thing – one I shipped last week was an agentic interface for Salesforce reports that understands their domain better (and all the undocumented tech debt) than the built-in Salesforce AI does and therefore manages the context better

It makes deployment for stuff easier for felix and co.. https://x.com/Etherealize_io/status/2042284300029345910

> I am also not sure who Stripe Atlas for.

It's for founders who don't have lawyers. My co-founder and I are both developers, we used Stripe Atlas to incorporate a C-Corp due to expecting to fundraise <1 year after incorporation. Stripe Atlas generates about 200 pages of legal boilerplate documents with very sane defaults so that your corporate structure, bylaws, IP protections, director indemnity, etc. align well with investor expectations. It helps investors not have to "rules-lawyer" all your corporate records during due-diligence, because their content exactly matches YC's expectations.

-------

I said we made a C-Corp but other founders should default to LLC, which Stripe Atlas can also streamline. An LLC is superior to C-Corp in pretty much every way for any pre-raise founders who don't have an extra $2,000 to >$10,000/year they're willing to part with for higher franchise taxes, "foreign" (different state) corporation registration, CPA's, and additionally lawyers if any investments aren't YC SAFE's (e.g. not YC, Neo, or A16Z SpeedRun).

Also note that for pre-revenue C-Corps, Delaware franchise taxes are scaled against number of shares, not company revenue or # of employees, so you can save some money by forming your company with 1,000,000 shares and then file a "Unanimous action of the board of directors" to increase it to 10,000,000 just before angel/pre-seed/seed round, and potentially save a few hundred dollars on your first year franchise taxes, depending on when you incorporate and raise. But if a few hundred dollars makes a difference to you, incorporating as an LLC instead of a C-Corp is the only defensible decision.

And as always, start your taxes 3-4 months before they're due. If you want a CPA to do them (which you should if you have any revenue), you'll need to retain them way ahead of time for C-Corps. If you're filling tax forms out yourself, you'll want to start at least a month before they're due.

> I am also not sure who Stripe Atlas for.

This was such a weird mention to see in the article. Stripe Atlas is a service that helps new businesses incorporate and onboard onto Stripe/partner services with some startup credits. It's been around forever, has nothing to do with AI, and is generally a very well-respected service.

I am curious. Say, cloudflare let's Mr doofus run 4 agents from AI provider X. Those agents go on to create a pyramid scheme, prompted or not(both cases are interesting to me). The agents are running code in infrastructure owned by a third party cloud provider. The law catches up with the bots many years later after people lose a couple million. Who is at fault?

I think cloudflare is in the clear. Mr doofus could argue that the AI company allowed or enabled the crime which they otherwise wouldn't have done. Or Mr doofus could claim his prompts shouldn't have lead to that outcome and that wasn't his intent at all. Making the bots at fault, but not the AI company I guess?

"Prove that you are a human" is such a nasty hack for lack of thought out solution for rate limiting. I'm really happy we are moving away from this.

Money talks.

There’s a completely free tier of Zohomail which does more than what I need for a custom email.

Perfect for spammers, scammers and domain squatters, who can now automate their activities even more.

Can’t think of any other uses for this given the current state of LLM ‘agents’, though I can’t wait for the next report of something like ‘openclaw registered 1000 domains for me without asking and now cloudflare won’t refund me’.

I’m not saying that you’re wrong.

But it’s worth noting that any good technology starts off being called a toy and with most people not being able to imagine its usefulness.

My biggest hesitation with these things is that there is no limit to the possible bill I may receive when the agent goes haywire. Cloudflare doesn’t see this as a problem of course.

That's pretty neat! What do you use to send and receive emails on your phone?

People use agents to deploy sites all the time. Buying a domain is part of that if you want to build a site that's beyond a toy. Allowing agents to do a task isn't just for things you do every day – it's also for things you do rarely and need agents' help. It's not just devs using agents to perform these sort of tasks anymore.

Stripe Atlas makes it massively easier for startups to incorporate in Delaware. This is particularly hard for non-US founders. It solves a real problem. I don't think this part will be done by agents though!

Disclaimer: I work at Cloudflare but not on this

This conflict is popping up everywhere. There is a push by a lot of companies to allow agentic use of their services (and new companies explicitly offering "X for agents"), ignoring the fact that "agent" means the same thing as "bot" which we've spent the last couple of decades actively filtering out. Will be interesting to see how it plays out.

I thought this was excessive and impossible, but as I was reading, I realized nowadays everything you say is technically possible. The future gives me the chills.

> By agreeing to these Terms, you represent and warrant to us: (i) that you have not previously been suspended or removed from the Websites and Online Services

CloudFlare ToS has you covered. A human must accept it, even with the new agentic flow.

Some would argue, forcefully at that, that AI cannot make art and/or cannot be used to make art.

What I saw was Transmetropolitan setup, where Hole renews their presence online every 5 minutes or so to avoid government censor.

Here's my top-secret Rube Goldberg Machine that maintains my online identity.

It's not as simple as being pro- or anti-bot. It's about giving site owners the tools to decide whether or not they want to allow them. Seems pretty consistent to me. If they don't want bots, they can use tools to identify and block them. If they do, they can do things like automatically deliver markdown versions to them, or use x402 to charge micropayments.

Disclaimer: I work at Cloudflare, but not on these

Like the Delamain AI in Cyberpunk. You would need to allow anonymous payments with cryptocurrencies for that, but it's coming for sure.

This is Cloudflare. They have an extremely strong incentive to increase bot usage. If there is no bot scrapping the internet they'll be out of business.

Can you support this claim with some evidence? Not just about the redundancies, but I’m also particularly interested in hard data showing Claude is capable of doing that kind of research with near 100% verifiable accuracy and migrations with no data loss and equivalent functionality (which is required to sustain your claim).

isn't cloudflare webworker and email forwarding infra hyperscaling and highly available?

Have you talked to Andreesen Horowitz yet? That elevator pitch alone should get you a few million.

Reminds me of an article from The Onion from this morning: https://theonion.com/taking-advantage-of-other-people-was-th...

About goddamn time. The recent past consisted of discord blocking me because their telemetry was broken and exceeded their rate limit and target blocking me because two devices in a single household look really suspicious.

I mean, Cloudflare will help website owners ban scrapers unless they pay. It’s kind of what they do.

yes but that's not good if you want programmatic inbound access which is what u need for many apps. That tier has no imap access.

The new goal is to flood the Internet with so much junk that human-created content can be sold back to us in a walled garden.

Yeah, like blockchain.

(Sorry for the snark, I'm hangry)

There's a whole payment section in the submitted article which addresses your concern, perhaps you should read it.

once you've emails stored, you can use any webclient.

you can write an api to imap adapter and use it in your favourite mail client

SES exposes SMPT directly.

Solid insight. What was once profane is now sacred. And it feels like it happened overnight. Lots to ponder on that.

The catalyst is probably the consent of payment processors, if I had to speculate.

In defense of old-school bots, we had to code them up by hand.

What about the 80% of teams? Are there enough trenches to dig in the country for them to make a living?

> Devops and sysadmins can no longer withhold information to maintain job security.

I can't imagine this is very prevalent. That's a very 2004-style corporate immaturity; I get the sense that even the slow-moving behemoths of the software world have mostly caught up to, say ... 2017's recognition of the importance of automation and reproducibility and won't tolerate the kind of malpractice you describe--wilful information siloing by infrastructure teams.

Like, those businesses might well suck at automation! But they've been doing it and firing the people who resist it for a long while now.

Epic. Can't wait for those humans to be rehired after you find out that letting Claude perform 1000s of migrations autonomously is a bad idea

Only downsized? I would expect them to cease to exist entirely in the coming years, as western companies begin to realize that AI is cheaper and more competent than the Indian firms they usually outsource work to.

And when it goes wrong, production is down, until they can get a real devops to look at what shit the AI-only guys did wrong. Haha, no serious shop would act like that, but then again most shops are not serious, now are they? So you might have a point.

You forgot "make no mistakes" and "don't hallucinate" and "don't delete any important files" as well, those are important.

I found that, without that, Claude makes too many critical mistakes.

> I wonder what it means for an agent to “agree” to terms of service

It's already not clear what it means for humans to do it, but it doesn't prevent every single service from asking it. At least an AI has a chance to ingest it all.

Wait until one account is banned, and then all linked accounts are permanently banned.

Maybe something like this?

AI agent calls a human on their phone (even engage in an email chain), whilst talking to the human they analyse the likelyhood of diffferent fraud vectors, and choose the most likely one to work on this particular victim. Whilst keeping the human talking in chit chat to raise their confidence levels, in the background it buys a domain which fits the users fraud profile, and quickly makes a basic website on it. Maybe its a fake login page, maybe it just hosts malware, who knows at this point. The agent then emails the user from a mailbox on the new domain which directs the user to the new domain and commits the fraud. The email from the domain ties up with what the agent is saying on the phone, so it all looks legit to the human. Immediately after the call it deletes the website, directs the new domains dns to blackhole and discards it from its posession.

This is all possible right now. I am also interested to see what is built with this technology in the future, but interested in a very worried way.

Yes I'm sure that whenever Andrej Karpathy complains, the market reacts. By the way, remind me who Andrej Karpathy is?

Well, let me guess... even more bots and new agentic spammers and scammers?

Interesting questions you bring up. Especially the legal ramifications as to how it would fully work within current legal framework. I suppose there would be a broad disclaimer and agreement one would have to agree to that would state that users of the service are ultimately responsible to monitor and ensure websites deployed by agents comply with local laws. Ultimately I assume that since it is not the agent who pays but a registered user that the user would own the site. And that the legal agreement would be agreed to beforehand so it is legally binding.

Buying the domain is the key here.

LLM generation in general provides the most use to scammers and the like. Generate emails which people won't read, generate articles which are just honeypots or rip-offs, generate images to said articles, generate more and more spam.

Every legit use case for LLM practically requires that human would verify the result manually, at least briefly. But spammers can enjoy skipping that step, since content was never a main priority in the first place.

And cloudflare can actually sell them priority access to pass their bot protection or introduce micropaiments for agents access content. I feel cloudflare is getting a bit scary tbh. It is like your friendly bot net.

The DNS provider I recently switched to surprised me with a policy:

To create records for more than one domain, you need to write a personal support email.

They say it's to raise DNSSEC awareness, but I think it's also a robot captcha.

It's not Cloudflare's job to see what you choose to buy as a problem.

Wouldn't it be critical if the agent botched the domain purchase in weird ways ?

Short of throwaway sites (spam etc) it's hard to imagine skimping time on this specific, mostly painless part.

> it's also for things you do rarely and need agents' help

I recently set up DNSSEC for the first time.

It really was just a bunch of copy-paste from one provider to another.

I like to understand what I'm doing, and LLMs helped greatly with that.

But it was copy-pasting screenshots into chat, so not really agentic.

Lets remind the purpose of incorporating in Delaware is legal tax evasion, so that we don't not have pensions, health insurance or anything nice, really.

Rename to Greedware.

The future is the internet will be entirely bot activity and humans will ether be strapped in to the metaverse reels ai slop feed or they will be outside interacting with people in person again. Both of these seem like likely futures and probably both at the same time.

I think this is just saying you can’t sign up for a new account after a previously created account gets suspended, not that the act of suspension itself causes you to violate the the terms of service in perpetuity because, pedantically, any suspension that has happened, happened “previously”.

The likely outcome is that the phone system becomes massively more locked down. Your phone will only ring if the caller has a number which is backed by a real ID, particularly one from your own country. It will become increasingly difficult to contact someone you don’t have a legitimate connection to.

The banking system will become increasingly fraud resilient with better real time detection of fraud.

Your phone may even have its own AI on your side listening in on the call and sounding the alarm when a number from Nigeria starts using an AI voice pretending to be your son.

People used to say the same about photography a while ago.

Are any of these domains public? I’d love to study and better understand the use case for needing to AIify this.

Humans will not win in court with a "but the agent did it, I had no idea" argument. Just look at how the cases against OAI are going, and that's where families lose a loved one. There's not going to be any sympathy when your agent committed fraud on your behalf.

And it's not like pro agent companies have a reason to self regulate. They're not going to absorb that liability voluntarily, they'll push it onto users contractually (most of them already do). This is just another channel to bring in customers. They will capitalize ruthlessly to increase their bottom line.

Cloudflare's prices are already close to unbeatable. They basically resell at cost. But there's nothing stopping you from doing that if you want.

is most sysadmins and devops capable of 100% verifiable accuracy? you over estimate average skill level available in market.

It's not about the uptime or scalability. Everyone has to make the choice for themselves if they value their time less than $12/year (Or free if Google is an option) for a critical part of their digital infrastructure to set all these moving parts up and keep them running over years.

I'll stick to Fastmail, where if something isn't working as expected I can just email them and get a response from a real human.

It doesn't change the fact that the workflow gp explains is a duct taped construct.

It's hyperscalable and highly available today, until the API changes.

A truly wonderful read

So true

That kind of captcha has a very short half life. Software ate the world now AI is eating software

It's the western great fire wall, good thing the things within the fire wall is huge and encapsulate still most of the world.

Are you perchance talking about deSEC? I've also switched to them, and thought that it was too much work to send an email and wait for replies, so I ended up using dummy inboxes for my other, lesser important domains.

Though I guess it's still a good thing they do this? At the time I remember being mildly inconvenienced, but not enough to actually care. I just remember thinking, "How is this nonprofit going to handle all that support volume?".

I'd assume they want to limit the number of bills that will get disputed.

Yeah it's highly available until it isn't and then that turns into your problem rather than something like Gmail just working

Curious, is there an Andreesen Horowitz Agent MCP?

Let’s automate this end to end, from idea to raising capitals. Vibe Angels should just be multi agents managing how much capitals to allocate to each projects.

I disagree frankly, as the next wave is clearly fully autonomous businesses.

This made me realise they’re doing the same thing the AI labs are doing: selling both the problem and the solution.

They are arming spammers and scammers with these tools so you need their product to protect yourself from them

I mean, Cloudflare was always kind of scary. They filter the world wide web, literally.

> To create records for more than one domain, you need to write a personal support email.

I'm not all familiar with this so I don't understand why it's not a ticket or any other non-automated action even for a single domain ?

I mean what is "the standard" that would actually allow a robot to register a domain to a DNS registry ?

If the rest of your deployment flow is via the agent, needing to switch over to a different context and open up a browser and login (or create an account) and buy the domain absolutely is a bump in the road.

People are skimping time in every part.

I am watching people who can't code build and deploy dashboards and sites with Claude Code (desktop app - they don't use the CLI), then go cap in hand to developer friends to get it hosted on a domain (rather than some Vercel or whatever URL).

Those people absolutely want to risk letting an agent buy and set up the domain.

This is not necessarily as blindly stupid as you might think. Many of these people know that this workflow is no good for writing code that does anything serious (i.e. storing data for people, taking payments, etc.) but there are a huge number of projects that are just websites, dashboard, data visualisations, etc. with static content and public APIs (Twitter is awash with them) and domains are cheap.

A decent minority of these are even quite cool or interesting.

So a lot of people want to put their vibe-coded weekend project behind a nice domain. Why not?

So pne step towards the Neuromancer universe.

The primary purpose of incorporating in Delaware is less about taxes and more that Delaware is the "Silicon Valley" of corporate law - incredible concentration of professionals, infrastructure, and intangibles. Any dispute you have will generally be handled better, faster, and cheaper by Delaware courts than they would be anywhere else. I'll quote my good friend who is a startup M&A lawyer: "I'd go so far as to say that it would be managerial malpractice to incorporate anywhere other than Delaware."

Last time (after years of doing it manually every once in a while) I just gave codex an ephemeral restricted Cloudflare API Token / key / whatever, the screenshot, and it set up all the records on its own.

Oh:D I am not saying they are right, but the sentiment has become rather strong lately.

> There's not going to be any sympathy when your agent committed fraud in your behalf.

Good thing the fraud is committed in places that specifically don't prosecute fraud when it's targeted against Western countries.

Cloudflare's prices are very beatable, especially if we're talking about the first year price.

Their name doesn't appear in the first 6 pages (~175 TLDs) of this list https://tldes.com/cheapest-domains

On renewals they appear much more competitive though.

Fraud requires intention

Investors usually expect that non-US founders incorporate in the US, and usually expect Delaware. There are other states that are more friendly to tax avoidance. Delaware is mostly preferred because it's a known quantity with mature regulation. Investors don't want to deal with dozens of different legal regimes, they want the one that they know about.

No, it’s not. Companies have to pay taxes where they operate regardless of what state they incorporated in.

Stop spreading populist internet bullshit.

Incorporating in Delaware is like 95% about being in a predictable legal framework for any business related dispute imaginable.

This reality also crystalized for me earlier this week when I saw a post about unchecked AI slop videos about WWE being posted to YouTube. Many of the videos suffer from the LLM stroking out (for lack of a better term) and devolving into mumbling, screaming and white noise. Yet, the comments are replete with obvious bot content which doesn't mention this at all and talks past the larger, flimsy narrative on display (i.e. AI-generated), anyways. We're exhausting our natural resources and reducing quality of life for a great number of real, live people so bots can talk past each other on YouTube.

So, if you're looking for me, I'll be hiking while it's still legal.

Also be aware most website ToS are worth the paper they're printed on.

You would just get called from an agent (bot) based in your country. There’s no easy way to prevent that. Fraud is massive and it’s becoming cheaper and easier to run at scale.

All of the domains are public. Whenever a new model comes out I like to ask a very specific prompt that helps me identify niche markets with high buyer urgency, have the AI rank them across a rubric, pick the one that has the highest degree of automation potential and then have it build me an MVP.

I’m not trying to shamelessly promote here but since you asked one of them is at jobwiz.biz

It’s not AIifying one thing. It’s AIifying the entire work flow… every detail. Allowing domain names is just one aspect of it.

The agent does everything. “Make a website that does…“ and it can handle everything from start to finish. It’s that good now.

You joke, but like the meme goes: go knocking on enough doors asking to see the devil, and eventually he might answer.

that's the thing it cannot stop working because webworker and email forwarding is very reliable, email itself has retries built it and soft bounce handling.

Yea, I appreciate them protecting it from DDOS. I always viewed them as a responsible company.

To me this feels irresponsible and like it's main goal is to forward autonomous cyber attacks. Which is antithetical to what they do? Maybe I am missing the legitimate use case here, but I can only see this being used for removing responsibility from crime or espionage?

Does anyone know offhand if cloudflare is a department of war contractor? I never looked into it. But this smells funny to me

Somehow the Internet needs biometrics and age verification everywhere but also chat bots can buy property there without too much thought.

[delayed]

Welcome to the grift economy.

You better mean “hiking” as in through the metaverse forest strapped into your corporate-sponsored VR headset, because outside time is for citizens only, friend.

Then when it gets reported the authorities can just look up the owner of that number and arrest them. Vs overseas based operations that are difficult to follow up on.

do you work on a cloudflare delaware-awareness project? Delawareness?

Your SSL cert needs to be rotated.

coming soon q2 2025?

Uhuh. And in other places, companies are incorporating in Ireland or Luxembourg or other similar tax evasion heavens because of the "predictable legal framework" too. Lol.

Perhaps more accurately they’re worth what it costs YOU in legal fees to defend them coming after you. Those are real dollars you still have to spend.

The question was what's in the dots. I have no doubt that agentic systems are good enough to buy domains and make one-shot websites from a prompt, but what is the legitimate use case for which you'd want to repeatedly perform "Make a website that does..." on a new domain?

Just a heads up I have seen complaints about CF email forwarding completely dropping emails that failed to pass certain SPF validation. They get completely dropped and the worker doesn't get called and they don't get forwarded, rather than in something like Gmail it would end up in spam

state did this for centuries or longer btw. they create violence/war in some area if the planet, then offer to "fix it".

like wmd's in iraq and hormuz problem now, lmao. remember how hormuz was not a problem and it was wideoy peaceful and open months ago? lol

Right, and in other countries they have different laws. In the USA they also pay taxes where they operate. That's how it works.

"Legitimate"? What scams are you implying are happening? A friend of mine wanted a site to help him sell DJ lessons. Another friend has a haircutting business that wanted a better site. Massage therapy. Etc.

That cuts both ways though. Nobody is coming after you unless it is worth their legal department's time (which cost much more than your own lawyer).