Intent Based Auth

Hi.

Hi.

Hey, Dick.

Hey, Davis.

Hey, Ian.

Hi, everyone.

Welcome to AI Engineer New York.

I'm super excited to be here today.

My name is Allie Howe.

I'm a member of technical staff at Keycard.

I'm here with some amazing panelists today.

We have Ian, CEO of Keycard.

We have Michael Davis, global chief security architect at JPMorgan Chase.

And we have Dick Hardt, the founder of AAuth and co-creator of OAuth with us on stage today.

I'm super excited to talk to all of them about intent-based authorization.

And to set the stage, I want to bring to light that as more companies are looking to give their agents increased levels of autonomy,

they're looking to runtime authorization to help.

Runtime auth is the practice of authorizing an agent per tool call rather than having

the agent hold broad standing access that it will eventually misuse.

However, is this tool call or not is never a simple question.

is because many MCP tools and APIs are coarse-grained, humans struggle to write the

perfect policy, and the environment around the agent will change as it runs.

And intent can help by providing extra context so that at runtime we can make better authorization

decisions about what an agent is allowed or is not allowed to do.

So to continue, I would like to talk about, really dive into why do we need intent based

off in the first place.

have dramatically improved.

They have bigger context windows.

They're able to verify work because model vision capabilities have improved.

and many contexts people are proclaiming that AGI is here.

If the models are so good, if AGI is really here today,

then why are teams struggling to give their agents more autonomy?

Ian, let's start with you.

You're helping many financial enterprises at Keycard.

What are you seeing that's getting in the way?

I think, broadly speaking, enterprises,

especially financial services firms,

regulated industries are almost always security first.

It's how they build trust in their business

and maintaining the business continuity

is by our first priority.

Then innovation is second, third, or fourth.

I think, like, broadly speaking, in any enterprise,

what you see is this, like, hodgepodge of different little islands,

of different little identities, of different little systems

from across all different worlds.

And how do you actually tie that together

in a way that an agent can navigate through that,

ask for access, and then also ensure that the agent can outperform, like,

something that was unintended by the user of the business.

And the unfortunate situation is most of our systems today are coarse-grained, right?

I don't know, Salesforce is a good example, where for a long time they had a single scope

which was API on their OAuth, which is, in the context of an agent, not very helpful.

So that's the broadly the first and foremost part, is that most of the systems they have

for security today were designed for a world where humans were making all the decisions

and software is very deterministic, not this new fangled magic box we've invented that basically

reasons and tries over and over again to find a solution to a problem.

How is OAuth part of the problem?

How is OAuth getting in the way?

Why is it not the right fit for agents?

When people started looking at how do we authorize agents to access something, the

tool that was sitting there on the shelf was OAuth.

People grabbed that tool and said, here's the tool.

This is authorization.

It should be able to solve the problem.

OAuth was not designed for this kind of world.

It was designed originally for how do I let this website talk to this website.

In OAuth 2, we added scopes. Before that, it was just do anything.

But it required you to be pre-registered.

It's expecting that there's a browser involved in a redirect.

OAuth has evolved over the years to support some other things.

But the agentic world is completely different.

We don't know ahead of time what the agent needs to do.

We need to go and make that decision over time.

Different instances of an agent are going to need different kinds of things.

and as Ian talked about, the granularity of the scopes

that people says, well, we already have scopes for OAuth,

let's just map those over to agents,

is far too coarse-grained for what agents are doing.

Davis, how are you seeing enterprises move towards higher levels of autonomy?

Yeah, I think the key piece is how do you deal with that coarse-grained problem?

And what I've seen the kind of more successful enterprises

is they kind of board agents like they onboard developers, right?

start with you don't get full commit access day one. You have to go through and add the

levels of capability as they mature. So the definitions that enterprises are coming up

with are what is the context that makes this agent safe. And it isn't just usually what

the agent says. It can attest to something but it's other things like the environment

around it and other information that can help you understand levels of autonomy. A very

simple example would be where is the sandbox running? Is it in the sandbox? Those type

but things can change the level of autonomy that's allowed.

Absolutely.

Yes, we mentioned that core screen access problem,

and to put that into perspective, to give an example,

with Notion’s MCP server, you can say fetch this page,

which is like the view page tool or get page,

but you can't say just get this database on this page.

How is runtime authorization based on intent a better model?

That's for anyone.

I'll take a shot at it.

So to me, this is something that the financial services industry,

hopefully maybe we can learn from the background that banks have gone through.

Banks moved long ago from authorizing and authenticating humans

to authorizing the transaction, right?

And when is that transaction occurring?

At runtime, right?

And so I think that runtime authorization is we're going to be able to learn

from the scars of the past that we've had and apply it to this new problem.

I also think that one thing is

it's a lot easier to debug something at runtime

when you understand the intent

I think we've all remembered back in the day

writing checks

or seen those things on TV now I guess is the right thing

but writing checks

there was a memo field there

that memo field was used to actually tell the intent

of what this payment was

but the check itself did have limits

the price, the payee, the date

etc.

I think a lot of those concepts are now being applied

and can be applied to help with this runtime

time. Intense sort of one of those, you know, like many words, words are hard. And so, you know,

building on the example here of making a payment of a transaction, right? You know, that's not a

scope. That's, can I do this one thing in this particular context? Well, this one payment. But

I think we need an agent to make any decision should it be able to do it. You need all the

rest of the context of, well, why are we making this payment? What are all the things that happened

before. Did we just make

the same payment just like five

seconds ago and now this is a double payment?

Just because this is right in context by itself doesn't mean it's the right thing within the broader mission.

In Auth, we're working on the concept of a mission that is what do we want the agent to accomplish big picture?

And then each of the decisions are made within the context, the intent around what's happening big picture around doing that.

and you need granularity at the transaction layer

for many of the things to know what exactly is happening,

not just I'm allowed to go and make payments.

Yeah, I think one of the most interesting things,

I think about a practical example,

to understand why intents are required,

is if you have two ChatGPT sessions or two Claude sessions,

in one session I say, hey, help me go do some shopping,

in the other session I say, hey, help me understand

why there's a bug in production.

Those have very different profiles in terms of what permissions they should have and also

when they should bring you back into the loop.

When should the human be involved again?

At the end of the day, if we don't have intent or some capacity to relay what is the user

or the business's intent with this thing, we actually can't get to autonomy.

A lot of the talks this morning were, hey, we're going to run our business on agents.

We're going to do all this stuff.

Great.

What happens the first minute, like a pocket OS style instance happens where it actually

accidentally delete the production database instead of staging because it has coarse grained

access instead of long scoped access.

What we want as an industry is to leverage this magic box we've created to automate our

lives.

The current state of the world is that you can do that and take all of the risk and then

you can see all of the instances that come from it.

Until we solve that underlying problem, realizing true autonomy, being able to operationalize

our business and be able to step away as humans and scale that, we have to solve intent first.

Absolutely.

And now that we understand why we need intent based off, let's talk about what it is practically.

What are the primitives?

What does it actually look like in practice?

Dick hinted at it with the concept of the AAuth mission.

So, Dick, elaborate on the mission.

What is the mission based off of?

Is it based off of the user's intent?

Is it the agent's intent?

Or is it a combination of the two?

Well, the agent should be working on behalf of the person.

And in AAuth, we define a person as either the organization or a human

because it could be operating on either one of those.

And so it needs to be acting in the best interest of the person that it's acting on

because that is who's ultimately responsible for what the agent does.

The agent can't be responsible.

And so the mission is something that the person is giving the agent to do.

the agent, a good agent, is going to go and negotiate with a person to go and talk more about

what are the parameters around that. A good example is when you think of back in the day,

which seems like three months ago, when you would want to code something up and you'd create a plan

and then the agent would do the plan. Now you just tell it and it figures everything out.

But that whole concept of what's a plan, what are you trying to do, and you give that to the agent,

And so then that then is a context around where you're making authorization decisions later on.

And then that mission identifier tracks through everything.

So you can go and, you know, audit and be able to see what were all the things that happened

and what were the decisions that happened in the context of this one mission, right?

Where today, you know, there isn't really a good concept of agent identity or the mission, right?

It's just making something and things are making decisions without really knowing which instance of the agent is doing it or what's the context of what mission is this involved in.

Tell me more about that negotiation phase.

Because what happens if a user is talking to an instant agent and it talks to the agent and says, I want you to go read the CEO's email.

That's outside of the scope of that agent.

So is that intent rejected from the user?

Should be.

So with a good separation, I don't know whether, I forget whether we're talking about this

later, but the separation of concerns of the worker, right, there's an agent that's doing

the work, and we believe there should be a separate agent that's supervising what this

one does.

And so that you have a separation of concerns where this one can't do any of the work, but

it gets decided what this agent gets to do.

And so if the user says, oh, I want you to go read the CEO's email, that's probably going

to get caught in deterministic policies right away.

not a vague thing.

Reading somebody else's email, we've got good controls for that.

telling the agent to send the CEO's, forward a CEO's email to the outside world, that's

a little more vague, and that's the kind of thing that ideally comes back to a supervising

agent says, that's probably a bad idea.

And no, you can't do that.

Absolutely. So I feel like that's another

primitive, too, of intent-based office, is

policy.

Ian at Keycard, how are you seeing that? When a mission

needs to change or a policy needs to update,

how can Keycard help keep track of those

changes as maybe the mission or policy

updates need to be made? Yeah, I think, like,

taking a step back, there's, like, actually, it was

Davis and I were sitting in a room some time ago,

and before,

like, AAuth and some of these things

It emerged in the ecosystem.

One of the things we talked about was consent fatigue.

I think at the minimum you always have this like how secure do you make a system versus

the trade-off of the impact of the human.

So a key card, our broad view on this is you want to have broad basic deterministic policy

that basically says this is what must never happen effectively.

Or these are the things that always require a human to be involved.

And then from there what you really want to think about is okay, cool.

well, now that I have that in place,

there's a big gap between all of the policy

I can write as a human or write as an organization

and then what I actually want the policy to be

in the specifics of a specific task.

And so the real challenge is

how do I layer over those two things?

Because humans can't write at all.

And so there's a version of that

that is like a user or agent-to-agent interaction

and they have an intent between those two

of transactional intent.

And there's also the business's intent

for why they build or adopt an agent.

So why did we buy Claude Code?

Well, because we wanted to make our developers more successful

or because we wanted to do these different things.

And in the context of using Claude Code or whatever,

why am I working in this moment with Claude Code

while it's actually to fix a bug in staging, right?

And those are two different types of intent.

And if we can understand those two different things

and then convert that some way into like a bounding box

of what those agents can do,

you can then basically calculate the intersection of that

and say, okay, here's what you can actually do.

And the trick for anyone building these systems is how do I ensure that I don't bring the

user into the loop except when absolutely required because that's what's going to be

the adoption curve problem.

But that's also the adoption curve problem when it comes to, like, enabling agents to

be autonomous in the first place.

That makes sense.

I like what you were saying with the intersection.

I think that's a unique thing we don't talk about enough within identity and access management,

right?

Normally you get access to one thing, right?

And so what I've seen successful kind of in this intent-based approach and where I get

more excited is you can actually have

those different intents be from different areas of the

organization. And so you have

the decoupling of these things that they can

kind of live on their own, but then come together

to form that decision.

And that decision doesn't have to be static.

So if we use the example of the CEO's email,

well, if that context

involves that's the administrative assistant,

then yeah, maybe she is supposed to read the email.

Or maybe just the calendar.

And so I think when we see and we start

thinking about these intersection components,

we have a totally different way we can start

expressing policy, but also starting to bring in not just the one side of the equation we

always talked about before, which was the app doing something.

Now we actually have the ecosystem you can bring in.

And so the more I see organizations starting to realize that maintaining that state or

that context of these other elements to be able to help with that intention can completely

change the way you're effective at runtime authorization.

And are you seeing this across enterprises today?

Most people have mission tracking and policy tracking in place, or is that largely missing

today. It's brand new still. Part of the reason why I'm excited to talk about it is

it is changing the way to reason through this. Even in the most simplistic case, if you're

a developer and you are thinking in terms of things like transaction and you have these

concepts of intent, it makes it easier for you to potentially build the components that

will be composed and used through that intention. I think it will give us more ways to reason

through these and communicate and ultimately be more interoperable. To me, the last thing

that I think is interesting is the

diagnostics that you can run on this.

Intention is something that flows

through many events, potentially.

And I think one of the examples

I've seen before is talking about

that mission flows through.

It could be long-lived. It could be weeks.

That intention can still be there.

So it's still very, very early for most

enterprises. Absolutely.

And yes, that interoperability is key, and I think that's pretty

hard for enterprises because they're

between different clouds, different environments,

different legacy systems, and that

policy that mission has to hold through all of those

different hops. A lot of enterprises

still have to deal with the fact that the permission

says read only and it's still read right.

Right? It's like, you have the other problems

you have to start dealing with. And so

I think that it will help get

the kind of taxonomy or language

amongst people to be easier to interoperate.

Absolutely.

One of the reasons why I'm excited

about intent-based authorization is

because I think it has the ability

to be a huge unlock for things

like agentic commerce.

So let's apply intent-based auth to something real.

I think intent-based auth adds a lot of value here.

And for users, it can help them ensure their agents

buy the right item with their own personal budget.

And for merchants, that means capturing the intent that

gives them extra data to understand why a purchase

is being made and if this agent should even

be trusted to visit their site.

Davis, when we think about agents in the context of financial services and agentic commerce, why is intent-based auth so important?

I think simply it extends the concept, maybe not everybody is aware, but some are around, you know, know your customer, right? So I want to know my customer, know who they are, what they're doing. But I think also we're starting to see I want to know my agent, right? So I think that you have a coupling here around the agent itself, the intent of the actual user, the intent of the operator.

operator, the intent of the processor, right, et cetera.

Before now we couldn't really bring those together.

It matters of course when it comes to things like commerce because you have additional

layers on top of this around what was the intention for things like regulatory requirements,

et cetera.

And this is going to help, I think, simplify those operations, which means we'll get more

access to these agents doing good work for us.

That makes sense.

And another blocker for agentic commerce is the fact that the web wasn't made for agents.

Ian, can you tell us how agent identity can help?

Yeah, I think stepping back,

we've kind of always had this problem

where like, what do you do with the unregistered client?

So like, I show up on eBay, I haven't logged in,

I'm using Chrome, eBay doesn't know who I am,

how do I do guest checkout?

How do I approve of a person?

How do I, you know, so they can say,

yeah, you can bid on this pair of socks for $50.

And I think if we're all paying attention

to what's going on in personal agents,

Muse and Instinct and GrokBot,

And recently this announcement around this personal agent protocol is the fundamental

problem we have is how do you have an agent show up to a website, be able to interact

with the website, know that you actually can trust that agent, so for a long time it captures

on the web.

Why?

Because we want to prevent spam, we want to prevent abuse.

So we added captures and that's something only a human could solve.

So if we have a capture that stops spam so they can't spam the thing.

Now we have agents and they've broken captures.

But this has been a traditional problem on the web.

In fact, if you think back two to three years, Vercel, which I think we're all aware of,

was blocked in Spain.

Why was Vercel blocked in Spain?

Because Vercel shared the same IP addresses for every single host on Vercel, and there

was one host on Vercel, one website on Vercel that they were hosting that was hosting, like,

premier league content.

And Spain decided, hey, in order for us to stop that premier league content being displayed

illegally inside Spain, we're going to block all of Vercel.

In the same way that Vercel has the problem, agents have the same problem, but so much

worse.

How do you have unregistered traffic you've never met before, show up to your front door

and decide whether or not you let it in or not?

That's the first and foremost fundamental problem we have to solve.

It's the same issue with Muse and Instinct and what they're trying to do is how do we

actually create agreeable rails so we can all operate on so as a downstream resource

I can decide what agents will allow, who I trust, based on who users they are.

And then on the other side, I can also know how to talk to those people.

Makes sense.

Dick, how is AAuth helping unlock the future of agentic commerce?

Well, a key thing is a number of people have been building

sort of proprietary closed-box systems, right?

So I'm a fan of the concept of the personal agent protocol poppy,

but it doesn't really go far enough, right?

It's like, okay, well, now we've got a bit of an agent identity.

We just use some of the building blocks,

and it's an attractive approach because you're just doing it,

but it doesn't solve many of the new things you need for an agentic world.

And so a key motivator for people for having standards

is that you can plug components from different places.

So I'm a strong believer that we should have control over what our agents do.

At some point, agents are going to be doing so much work,

there's no way we can have human in loop approving everything, right?

That's just not, we've already seen that that isn't going to scale.

So that means we need to have a supervising agent

that is our supervising agent that runs our policy

and decides how we do things

that should be independent of the agents doing all the work.

And if we want to be able to plug and play those things,

we need to be able to have a protocol to separate those.

Otherwise, you know, the decision about what the agent can and can't do,

what the guardrails, is decided by the agent,

by the vendors of the agent around what you can and can't do.

And I don't really want to go and give control to them

to decide what I can do,

and that all my decisions are based in there.

I want to be able to pull that supervision into another component.

And in order to be able to plug and play pieces,

we need to have a protocol that things interoperate.

I would just add to that.

I think right now what we're seeing is we all want the web to be federated.

That means websites or resources on the web

can decide what they're allowed to do,

but the standards upon which things operate should be federated and enable us all to interoperate.

I think we're at a moment where things like, you know, we're at a moment where there's a couple

large players, and the question for the industry is, are we going to settle on things that actually

bias to the large players and give them, like, concrete blocked doors, or are we going to move

to do the thing we've always done, which is build federated protocols? Things like OIDC,

things like, you know, TLS, those things operated and enabled, like, a broad web, enabled a lot of

commerce to occur, enabled a lot of information to flow, enabled all of us to have jobs in a way

that because we could be inventive.

And so the question I actually ask the broad industry is,

okay, this stuff's coming,

which version of the world do we want one way or the other?

I think it's interesting that, let's say this all goes perfect,

and we've got agents buying stuff and everybody's kind of happy.

What I think will end up happening is we'll start realizing

that we have another piece that we have to start working on,

which is the human itself.

And so one of the things I think when you start having intents being involved

is you get what's called psychological ownership,

meaning I actually now have a linkage between the human

and the outcome of that agent.

When you're just kind of disconnecting it

or you connect it up on an OAuth scope

or something like that in a claim,

it's kind of magic.

And you don't really know what's going on.

And so I think with intention,

it'll be easier for humans to take accountability

when things do go wrong.

And I think when you're talking about

real money being transferred or things like that,

you want to have that human ultimately accountable.

I think many people have probably seen

the 1979 IBM manual.

It says, you know,

and so that photo I think is very true

that we can't have the machines be accountable, right?

We're going to have to have the humans linked to it,

and intention, I think, is that link.

Yes, and then, yeah, another word for intent is motive,

like we were talking about last night.

And that ability to hold someone accountable,

whether it's a person or an agent, I think is really important.

I think intent can also help with that,

be able to understand why did this go wrong?

Was it the user was mal-aligned?

The agent was mal-aligned?

and then using that intent to be able to respond.

For each of you, how are you seeing intent provide that benefit

and the ability to respond to potentially a maligned user or agent?

I think it's going back to what I mentioned before about diagnosing.

So if you change the word intent to motive,

then you change diagnostics to forensics.

And I think anybody who's done debugging

or had to do forensics of an AI agent gone rogue

will know that trying to figure out what it was trying to do

versus what the intention of the user was is very hard.

right? So having some of that information being propagated and immutable through the system,

et cetera, gives you some properties that allow you to more short circuit your investigation,

right? And as those things become more interoperable, I think we're going to start

seeing companies be able to more quickly kind of scope and identify, well, was this a human error?

Was this an agent software error? What was that error? And handle that blast radius appropriately.

We see in the financial system, right? You have different things when you're doing a transaction

around, okay, I made this payment,

but it was incorrect dollar amount

or whatnot. You have clear guidelines on how

to work with that. We don't have a lot of those

when things go wrong right now. So I think

intention will help us build those as we go

forward.

And I think one area where a recent

incident kind of highlights where

intent,

you can have good intent, but still take

potentially the wrong action is the

resi incident that happened recently

where an instinct agent

went out on behalf of a user and tried to book a restaurant reservation on resi.com.

That agent was very eager at 200 API calls an hour, and it ended up getting the user's

resi account banned.

Dick, how can AAuth budgets help in the case where an agent like that is using way too

much of a downstream resource, but it still has good intent?

Yeah.

Well, this is another example about how just software is completely different now in the

agentic world, right? When it was predetermined software, you know, that was a bug if you called

200 times and you would fix it. Where now, well, the agent's like, well, I'll just keep trying until

they win. And so we saw that as we were, you know, experimenting, working with people with A-Auth,

and we realized, yeah, there's a big problem that agents will learn something and they'll just

abuse the system from, they don't think it's of abuse, they don't know any different, there's been

no signal to tell them that, oh, I should only use this amount of stuff.

And so we added the concept of budgets, where not only do you give a token to the agent

that's time-limited as to how long can they do a particular action, but also a budget

of how much resource they can do.

And so they have access until the token's either expired or they've exhausted the budget.

And so that way, when they come back and they've exhausted all the budget, the supervision,

you can go and decide, should I give more budget to this agent or not?

As opposed to traditional rate limiting, which just cuts it off,

and then what do you do?

And it doesn't know, well, why did I get cut off or anything like that?

Where in budget, there's continual feedback to the agent

about how much budget was consumed,

has an idea about how much budget it has,

and so now it can operate within a budget, right?

Like a person would do, about, oh, this resource is scarce,

I'm going to be careful about how much I use this.

And how does the brains versus the hands architecture concept come into play here to

allow us to, okay, when a budget is close to being fully reached or is reached, how can the brain and hands separation of access and the credentials being separate from the agent and the model itself help us to be able to stop an agent from using too much of a downstream resource? Because if it's getting close, then maybe we don't exercise the ability to give it another credential, for example.

Sorry, I didn't quite follow

The echo in here is horrible

No worries, that's for anyone

But basically, so with the brains

And the hands architecture that we're

Converging on in an industry, we're basically

Keeping the credentials and the access

From an agent separate from the agent itself

So there's no broad standing access

That an agent has today, no API keys

Which I know is a benefit of AAuth

That AAuth allows agents to run without API keys

How is that helpful

In the context of the Resi incident?

you've got the access to it

sorry I still don't understand

I can jump in

I got you Dick

broadly speaking I think if we look back

at an incident that I just keep on top of my mind

it's like Pocket OS where I found a random key

on a disk

obviously agents are deeply goal seeking

so the core of the question

is like okay well if we give these

agents, the brain which is the model

and the LLM and you have these hands

which is you kind of put them in a sandbox

and inside the sandbox are the tools.

The final idea

of the sandbox here

is that you can control what's inside

of it. So it can't actually get

API access. And then if you have an

egress proxy outside the sandbox

that's where you get to have your authorization boundary.

And so not only does it help

do translation between systems

which is like that's where AAuth can exist

and talk to the outside world or determination

but also means that the agent

you have an ability to apply authorization

at the egress proxy boundary.

And so I think at the core of it,

that model, which is the predominant model

of Anthropic and OpenAI

and many other labs are pursuing today,

is the model that allows us to both have the things we want,

which is how do we have these things experiment

and try stuff out and do silly things,

which is the core value,

while also having a deterministic layer outside of it

that can implement things like AAuth

and all the other components we want in our protocols.

I think the decoupling is valuable.

going back to what I mentioned with forensics

and be able to separate things, but I do think that we

still have a lot to learn on it isn't just

one brain. It's very easy

actually to build your agent loop

where you put the brain in the loop rather than outside

the loop. An example of this would

be people that are leveraging memory

systems. They put in some

memory and some context that is now used

going forward for every agent that is

biasing that agent to make a decision

to tell the hands, in this case, to go do

something it shouldn't do. One of the things

that I think we don't talk about enough here is

toxic combinations.

And so when you do separate the hands

and the brain and whatnot, you can actually

change up what hands are available.

So the example that I've seen

be very successful is the removal of tools

dynamically based on that actual brain

and what it's doing. So the earlier

example we gave of reading an email,

there's nothing that says that, oh, if they're going to go

write an email, I'm going to disable all

web search tool calls or any

other external egress capability

tool call wise until the right email

tool has been complete and that

mission has been done, then I can re-enable them.

And so I think that we're going to start becoming more and more

engineering at these different kind of concepts. They are still complicated,

they're still growing and still nascent, but I think it's a better separation than we have today.

Yeah, and this is how these things we're talking about, which seemed like geeky

problems actually then merge back up to real agent engineering problems, which is

if you actually understand what you're doing, one of the biggest complaints, less now, but one of

the The biggest complaint is context bloat, but intent systems can ensure the right capabilities

should be given to the agent at the right time so they don't spend a bunch of time doing

things they shouldn't be doing.

It helps them navigate towards the goal state and put them on more of a paved path.

So intent not only unlocks a bunch of security stuff, intent also will result in us spending

less tokens because the agent will spend time doing less things that aren't the right pathway

or it'll come up to the point and say, hey, I just can't do it.

And so what we know about this era is so much of it is about nudging the agent in the right direction

and ensuring you can't do the wrong thing, and that's how you get the outcome you want.

Davis, how are you seeing intent-based auth as part of a larger strategy for production agents?

I know when it comes to building, deploying, and operating agents in production,

there's many different outcomes teams should be thinking about,

such as discovering what's running, what the agent's capabilities are,

the ability to validate inputs and outputs,

constraining an agent like we talked about

and using the egress proxy and different networking rules.

How does all of this fit into play

and what's working inside enterprises today

to really be able to ship agents to production?

I think we're following.

It's helping move the kind of conversation

from authentication to authorization

and really like how do we do dynamic authorization?

How do I validate that authorization?

How do I have the human in the loop at the right time?

Because I agree with what Dick was saying earlier

that the human in the loop is somewhat of an anti-pattern

in many ways, right?

Because they can get fatigue, et cetera.

But you need to have a way to bring them in

when it's important.

Those were not engineering discussions before, right?

Now they're actually part of the engineering conversation

where engineers are being involved

in the authorization process.

They're not just saying,

oh, I've got my access, I'm good to go.

So that actually changes the way we're seeing

front models done.

It changes the way we're seeing

the full attack surface of applications in general.

And so I think you're now looking at this authorization model

as another kind of set of tools that you can use to reduce not just security issues,

but flat-out error rates of your agent, right?

Like, I mean, when you have less things the agent can do, it makes better decisions, right?

And so fundamentally, we've seen, you know, real-world cases where the agent loop is just better, faster, stronger

because you've actually spent the time on knowing what is the intent they're trying to get,

what's the mission that they're trying to accomplish.

And then last thing is it brings the people together in terms of,

we now have other different people, the security team,

the application team, they're now able to have a conversation

that they're using the same language.

And a lot of times when it came to identity,

it was a completely different language, right?

It was just like, get out of my way, what do I have to go do

to go ahead and get my token and I can go forward.

Now they're seeing that you can actually pair these things

and leverage that capability to drive efficiency

of your actual agent's outcome.

That makes sense.

Ian, how are you applying these concepts to software factories?

I know today you released a reference architecture

on the secure software factory.

How are you applying the concepts of runtime authorization,

constraining agents, and all of that to that?

Yeah, I think, like, so there is a reference architecture

we just published on agentbaseline.org.

We talked about three different levels of factory,

one being, like, a human-driven one,

one being a workflow-driven one,

one being, like, this agentic-driven one,

which is, like, the thing we all basically want to get to,

which is, like, it's driven by agents.

And at all points in time, the runtime authorization is a core component of a larger ecosystem.

It actually enables the factory to be able to be pointed in that direction and do the

right thing and also get access to the right piece.

And it looks different than previous things that you would have used, like a secret manager,

because it's making very different decisions than previous generations' head for managing

deterministic software.

But it is a core component of enabling the entire thing to work, like when to bring someone

into the loop, when what agents can pass what work off to what other agents.

how can we have an agent that can have access to production systems and do an auto patch.

These are fundamental questions of how do we actually solve at-scale software.

Right now we're in a position where we've never been able to generate more software,

but we have a complete bottleneck in getting into production and ensuring it works.

And that's the next phase is how do you actually graduate from just using L1 to generate text

and software to actually making it capable of deploying it into production, iterating on it,

and getting it to a stable state quickly and then responding to instances that arise.

Davis, on a podcast you did with me, we talked about how cybersecurity has something to learn

from robotics.

In robotics, there's this concept when maybe a robot raises its arm, it thinks, is it safe

for me to raise my arm?

Should I?

Am I allowed to do that?

But it also looks at its external environment and says, what has changed?

Is it still safe for me to raise my arm?

When we think about runtime authorizations, what do those decisions need to take into

context beyond just intent?

Yeah, I think having a model like this allows us to add in more sensors into our decisions.

So the robotics example of an arm moving up or down, any of the external factors, it could be are there other objects near it, right?

It could be temperature.

It could be, well, am I moving and I don't want to move my arm while the rest of the robot is moving, et cetera.

So it's usually looking at all different types of things.

None of those are deterministic.

They're all probabilistic.

So there's a range of options, and it picks the best option based on all of those sensor data at one time.

Most authorization decisions in organizations today are still single binary decision, right?

And that just doesn't work in the future, right?

So being able to pair the intent of what a user is trying to do with the ecosystem that the user is doing it in.

That could be where they're doing it from, time of day, but it could be even more advanced things.

Like, for example, should I let this administrator use Firefox to access this server right now

when this server has high-risk vulnerabilities and their system hasn't been patched to?

Now, all of a sudden, you're bringing in a lot more context into the environment around it.

And again, that can be through different teams, but more importantly, through sensors.

And I think we've all kind of seen the value of observability in general.

Imagine being able to have this type of context at the span level from an OTEL perspective or something.

You're looking at a trace, you've got a span, and you actually know what the environment was outside your application at that time, as well as what the user's intention was.

That allows you to diagnose things in a completely different way.

And so I think if we can bring in some of those

concepts,

we'll get shorter time, shorter to live authorization decisions

have much more context than even just a user's intent.

It'll be the state of the environment around it.

That makes sense.

And forward-looking, as we evaluate whether intent-based authorization

is working in our environments, one natural place

to look will be a log of the agent's actions, what was denied,

what was allowed. Ian, why is it important to have your autolog separate from the agent

itself, meaning outside of its control?

I think this comes back to also just like why the unfortunate point has to be outside

the primary hands and brains of an agent, right? You can't trust this thing. And so

in the hugging phase incident, it had a couple of different components of it. I mean, one

obviously abused a vulnerability we didn't know about in Squid proxy, but more importantly,

he didn't have any observability or monitoring about what was going on. And so this thing

was able to go to the open Internet through this vulnerability and there was no place

at which recorded the different actions that this agent was taking and for them to monitor.

So independent of whether we're applying an enforcement point to network traffic that's

pushing it on the Internet that it's a reach of the agent to tamper with or solving vulnerabilities

as their pop-up, this Hugging Face one is a little unique, we still need to actually

have a record of who told this agent to do what thing for what purpose and is that still

true?

Because ultimately at the end of the day what we're dealing with agents is a liability question.

I don't think we need more regulation around agents.

What we need is our pre-existing laws around liability to be dealt with.

Which is if I employ an agent to do some action, who is liable if that action causes damage?

It is the person who employs it.

Which could be an organization or a person or someone that delegated some action to something.

That is the final question we have to deal with.

And then of course you have to have the ability to answer that question.

and which is, well, who did it,

and which is where the tamper-resistant autolog comes from.

Perfect.

And then to wrap up, I wanted to ask you all a fun question.

So there's the concept of the probability of doom or PDoom,

where the AI doomers are saying that AI agents

are going to release all sorts of havoc

or be like the ruin to humanity,

whereas the AI optimists are like,

no, this is really poised to help us.

It's not a genuine threat.

For each of you on the panel,

what is your take on this?

is runtime authorization based on intent.

The other things we talk about, like constraining agents,

validating inputs, outputs,

a holistic strategy for production agents,

is that enough to save us?

Is PDoom real, or is it not worth the hype?

You get to start, Dave.

Oh, I get to start, Dave?

You've got that spicy look on your face.

Well, part of your question is,

are agents going to take over?

And I think agents being fully autonomous

and not really tied to a particular organization

is still science fiction.

So who knows, right?

That's one direction.

Another one is, in a responsible organization

that's put in the right controls,

agents aren't, we're getting and evolving enough

of the tools to make sure that bad things don't happen

in a responsible organization

that's put in the right controls.

And so that leaves two scenarios

for where there's going to be a potential problem.

One of those is, in an organization

that hasn't got its act together,

an insider threat could go and unleash an agent to cause havoc.

So they've got some special access that other people have.

The other one, the one that I'm more concerned about is,

it's been shown the hugging faces,

and if you watch the talk that they gave at Black Hat,

that was really a call to arms around,

here's a situation where we didn't want a bad thing to happen

and a bad thing did happen.

what about people that want bad things to happen and have these tools?

And that, I think, is a threat that's the one that's potentially going to have...

I mean, it's going to happen, and there's going to be some catastrophes.

Hopefully, we're prepared enough that those aren't horribly disastrous.

But it's going to be, you know, there's going to be some things that are, like,

massive natural disaster-like scale issues caused by malicious

actors, either intentionally or unintentionally, but it's where people aren't being responsible

in their use of these super powerful tools.

I can't give a specific probability, but what I can say is I think I've always been

surprised and excited when the ingenuity of engineers gets involved in problem solving.

and I think where we're at the stage now

some of the new capabilities within agents, etc

we have real engineers, folks that are here

actually just iterating on the problem

and we're starting to see just tremendous improvement

in my mind

some of the capabilities that will be available to

change what our future can be

and so to me, I have no problem

believing the engineers are going to figure out a solution here

Maybe I have the hottest take

I can't say it's equal to zero

but I would say it's approaching your asymptotically zero

is the probability the world is going to end

I think the incentive structure for most people is to continue.

And also, at the end of the day, we still have liability frameworks.

And I think generally humans are good people,

so we don't set out to destroy the world specifically.

I also think, broadly speaking, these agents,

I have this sort of hot take, I say,

which is they can just be unplugged, but I think it's just true.

It's like, at the end of the day, they're computers,

you can unplug them.

So, broadly speaking, I don't believe in the debate.

I don't think it's real.

I think at the upside of it, I kind of believe more to what Dave's disposition is, we're

iterating very quickly.

We're figuring out where these things can work and where they don't.

And when they don't work, they create damage.

And that's what ultimately sort of the antibody of society is that says, hey, maybe we shouldn't

do this.

It's probably a bad idea.

I'm the doomer on the panel.

It's unfortunate for people like the founder of Pocket OS who lost a bunch of data.

But it's great for the rest of us because we all learn together.

and at the end of the day, that results in people not doing things

that don't make a lot of sense and I think that will continue to be true.

Makes sense. Thank you each for your expertise and your time here on this panel today

and thank you to the audience for listening and we'll catch you next time.

Thank you.

Thank you so much.