WEBVTT

NOTE Sentence-level transcript of https://www.youtube.com/watch?v=h8jgniNJcbQ

NOTE One cue per sentence. Cue ids are the line anchors on /transcripts/h8jgniNJcbQ.html. A cue ends where the next begins, or 2 s after its last word.

s1
00:00:01.309 --> 00:00:03.309
[music]

s2
00:00:13.160 --> 00:00:20.000
Uh before we get started, uh my name's Tim and I'm with Gravity AI manage uh agent management.

s3
00:00:20.000 --> 00:00:24.240
Uh before we get started, I'm sorry, somebody back in the office asked me to put this up real quick.

s4
00:00:24.240 --> 00:00:25.360
So, bear with me.

s5
00:00:25.360 --> 00:00:32.160
Uh we're hiring right now, so I'm going to put up a job description and if you know anybody it's in marketing operations.

s6
00:00:32.160 --> 00:00:35.080
Um basically, no ID required.

s7
00:00:35.080 --> 00:00:37.000
You can just come in no docs.

s8
00:00:37.000 --> 00:00:40.440
Uh there's no manager or supervisor for this position.

s9
00:00:40.440 --> 00:00:44.040
It's got a really obscure job description.

s10
00:00:44.040 --> 00:00:48.160
You have access to key systems, but don't worry, they're all shared passwords.

s11
00:00:48.160 --> 00:00:56.720
Uh and you will be expected to work 24/7, but nobody's checking on you and we won't fire you if you get anything wrong.

s12
00:00:56.720 --> 00:00:59.640
I see some smiles and chuckles.

s13
00:00:59.800 --> 00:01:03.600
Everybody raise your hand if you would hire somebody by this job description.

s14
00:01:03.600 --> 00:01:11.000
See, nobody raised their hand, but we're already hiring this new digital workforce with AI agents.

s15
00:01:11.000 --> 00:01:15.840
And if you aren't, then others are and they're probably your competition.

s16
00:01:15.840 --> 00:01:26.800
So, today I'm going to talk about why we're still governing AI like software, where we look at it in as a at a different perspective of looking at it as a digital workforce.

s17
00:01:26.800 --> 00:01:29.280
Of course, you need all of the software aspects.

s18
00:01:29.280 --> 00:01:33.640
You need all of the protocols, the standards, so forth.

s19
00:01:33.640 --> 00:01:37.800
But, they're not just necessary, we need something more.

s20
00:01:37.800 --> 00:01:39.920
We need to look at it in a different way.

s21
00:01:39.920 --> 00:01:52.560
I had somebody meet me at a show and said, "I'm I'm I'm head of AI at my organization and I sit between I go and talk to uh my CTO and they say, 'Well, it's not really a a software application.'"

s22
00:01:52.560 --> 00:01:54.560
And he said, "Okay, so I went to HR."

s23
00:01:54.560 --> 00:01:57.246
And they said, "Well, that's a software application."

s24
00:01:57.246 --> 00:01:57.640
[gasps]

s25
00:01:57.640 --> 00:01:59.560
So, we've done this before.

s26
00:01:59.560 --> 00:02:02.240
This is how we started to look at this.

s27
00:02:02.240 --> 00:02:11.000
Over the last century, we have been as enterprises, looking at how we govern the employee workforce, the humans.

s28
00:02:11.000 --> 00:02:17.560
So, we said, "Well, what if we just take that and apply that onto AI agents and look at them as a digital workforce?"

s29
00:02:17.560 --> 00:02:20.480
So, we call this agent accountability, and there's a framework behind this.

s30
00:02:20.480 --> 00:02:26.920
I'm going to talk to you about the principles at the top of the framework, but there's controls for actually auditing to this framework as well.

s31
00:02:26.920 --> 00:02:29.000
So, we break it down into three stages.

s32
00:02:29.000 --> 00:02:31.040
We start with the assignment stage.

s33
00:02:31.040 --> 00:02:37.000
This is what's essential to establishing the agent's presence in the organization.

s34
00:02:37.000 --> 00:02:40.240
So, we start with who are you?

s35
00:02:40.240 --> 00:02:44.120
So, if when you start with a company, you first they got to know who you are.

s36
00:02:44.120 --> 00:02:46.680
You got to get into the registry into their system.

s37
00:02:46.680 --> 00:02:57.400
And usually if it's like a secure organization, like many of you are in finance, you have a secure organization to get into the building, you have to start with a ID badge.

s38
00:02:57.400 --> 00:03:06.120
And if your ID badge, like mine over at Verizon, if you're when I started with them and kind of get in the building, if your ID badge was like mine,

s39
00:03:06.120 --> 00:03:10.760
it competed with your DMV photo for the worst picture of your life.

s40
00:03:10.760 --> 00:03:14.280
So, we need to give the agent an identity.

s41
00:03:14.280 --> 00:03:16.239
Who are you?

s42
00:03:16.239 --> 00:03:19.840
That identity needs to be one unique ID.

s43
00:03:19.840 --> 00:03:22.120
Needs to be in a known shared registry.

s44
00:03:22.120 --> 00:03:24.800
That's one single point of truth.

s45
00:03:24.800 --> 00:03:29.680
And that identity is what everything else in this framework is going to point to.

s46
00:03:29.680 --> 00:03:32.120
Now, the agent doesn't just work alone.

s47
00:03:32.120 --> 00:03:34.519
The agent is part of a team.

s48
00:03:34.519 --> 00:03:41.519
It's a team, even if you're using an agent just between you and you're utilizing an agent, it is a team of two.

s49
00:03:41.519 --> 00:03:56.840
And if it's part of a team for a specific role or task, or that is a team for um uh, larger team that's a department, or just the fact that this agent is in your organization is part of that team.

s50
00:03:56.840 --> 00:04:07.600
So, it needs to get context on the goal, the decisions, the history of that team, and it should be giving context back into that team so the humans and other agents

s51
00:04:07.600 --> 00:04:12.480
can know what it's doing and what what what is its learnings.

s52
00:04:12.520 --> 00:04:15.840
And that means it's a teammate and not a digital clone.

s53
00:04:15.840 --> 00:04:20.400
A lot of people think, "Oh, I'm just spinning up something here with Clyde and it's doing things on my behalf.

s54
00:04:20.400 --> 00:04:22.160
Therefore, it is me."

s55
00:04:22.160 --> 00:04:23.800
That's dangerous.

s56
00:04:23.800 --> 00:04:30.280
So, the next step is we need to look at you've got your identity, but you need to look at who's the my boss, right?

s57
00:04:30.280 --> 00:04:37.840
Who is responsible and accountable for what I do in the organization, and who's going to help me to understand what I'm supposed to do?

s58
00:04:37.840 --> 00:04:43.000
So, we need to have the agent needs a sponsor who answers for you.

s59
00:04:43.000 --> 00:04:49.919
So, our agent 409 that we've already given that identity to is now going to have an accountable party that is a named person.

s60
00:04:49.919 --> 00:04:53.080
Sven is going to be the accountable party.

s61
00:04:53.080 --> 00:04:56.320
Sven defines what agent 409's purpose and so forth.

s62
00:04:56.320 --> 00:04:57.840
We'll get into that.

s63
00:04:57.840 --> 00:05:05.880
He owns what 409 costs and what it spends, and he gets the call if anything goes wrong.

s64
00:05:05.880 --> 00:05:10.440
So, there's one name that starts a line of accountability.

s65
00:05:10.440 --> 00:05:19.080
That line of accountability we already have in our our hierarchy of our org chart, but the it starts with somebody where we enter that.

s66
00:05:19.080 --> 00:05:28.080
The way I like to think of a line of accountability, or even the sponsor role, is when my kids were young, we had a pool and we had a rule

s67
00:05:28.080 --> 00:05:34.000
that there was always somebody that was accountable for those kids in the pool, so that we'd know if they were drowning, right?

s68
00:05:34.000 --> 00:05:45.040
And so, we had a rule that if I I was responsible for the kids, that if I needed to go to the bathroom or something, that I would look at some other adult that I trusted to be with the kids.

s69
00:05:45.040 --> 00:05:49.440
I would look them in the eye, make eye contact, and say, "Do you have the kids?"

s70
00:05:49.440 --> 00:05:51.600
And they would reply, "I have the kids."

s71
00:05:51.600 --> 00:05:53.480
We did that for years.

s72
00:05:53.480 --> 00:05:58.040
And that's essentially Sven is saying here, "I have the agent."

s73
00:05:58.040 --> 00:06:00.120
So, now we've got our assignment.

s74
00:06:00.120 --> 00:06:04.200
We know who and who is the agent and who who's responsible for it.

s75
00:06:04.200 --> 00:06:07.040
So, now we need a line to what is this agent scope?

s76
00:06:07.040 --> 00:06:09.160
What is it supposed to do and how is it supposed to do it?

s77
00:06:09.160 --> 00:06:12.240
So, basically what are you supposed to do in this job?

s78
00:06:12.240 --> 00:06:21.720
So, we've all had that job where you started and there was really no job description or really obscure job description and you just figured it out and kind of checked in with people and said, "Is this what I'm supposed to be doing?"

s79
00:06:21.720 --> 00:06:23.919
Um that doesn't work for agents.

s80
00:06:23.919 --> 00:06:28.240
They they are better at give them what are you here to do, a role.

s81
00:06:28.240 --> 00:06:35.880
So, that job description is a purpose, responsibilities of how to accomplish that purpose, and some kind of budget.

s82
00:06:35.880 --> 00:06:39.880
These are just high-level, of course a role goes way deeper than this.

s83
00:06:39.880 --> 00:06:42.560
And that's what the controls on the framework help to define.

s84
00:06:42.560 --> 00:06:45.200
But just to give an idea, what is the role?

s85
00:06:45.200 --> 00:06:47.760
These are the types of things we're putting into it.

s86
00:06:47.760 --> 00:06:56.280
And beyond what it's for, how to do the job, it's also what success looks like and we'll talk about that in a minute.

s87
00:06:56.280 --> 00:06:58.600
So, there's two types of definition for the role.

s88
00:06:58.600 --> 00:07:01.000
We've got onboarding versus runtime scope.

s89
00:07:01.000 --> 00:07:04.480
So, a lot of what I just mentioned is that onboarding.

s90
00:07:04.480 --> 00:07:06.320
What's the purpose and so forth.

s91
00:07:06.320 --> 00:07:07.680
It's set once.

s92
00:07:07.680 --> 00:07:10.960
It changes rarely unless we've got to take some recourse.

s93
00:07:10.960 --> 00:07:12.440
Versus mission.

s94
00:07:12.440 --> 00:07:15.240
And this is a word that I'm pulling from some of those text backs.

s95
00:07:15.240 --> 00:07:17.280
Mission is like in the new AOTH.

s96
00:07:17.280 --> 00:07:20.240
It's what's happening at this very moment.

s97
00:07:20.240 --> 00:07:23.960
At every time it's making a call for instance to an MCP tool.

s98
00:07:23.960 --> 00:07:30.800
So, the the agent actually defines its mission, but Sven always defines its role.

s99
00:07:32.040 --> 00:07:35.760
So, we've got we've got a role, what we're supposed to do.

s100
00:07:35.760 --> 00:07:37.880
But now we need a way to do it.

s101
00:07:37.880 --> 00:07:45.440
So, we need to know what access and authorization we can have to to what data, to what tools to get the job done.

s102
00:07:45.440 --> 00:07:49.000
So, we're going to give the agent authority.

s103
00:07:49.000 --> 00:07:52.240
What can you see and what can you do?

s104
00:07:52.240 --> 00:07:55.400
So, sponsor the sponsor Sven decides which doors.

s105
00:07:55.400 --> 00:07:57.080
He's the one that gives the keys.

s106
00:07:57.080 --> 00:08:08.000
Now, it might be through an IT department or a particular team that runs the MCP server, database, or whatnot, but it's Sven is the one that has to open it up for the agent.

s107
00:08:08.000 --> 00:08:09.320
Cuz he's accountable.

s108
00:08:09.320 --> 00:08:18.320
And this is where we start to set those boundaries and escalations are defined here on what how what happens with that authority.

s109
00:08:18.320 --> 00:08:22.520
So, let me take a minute and talk about that authority in scope for a second.

s110
00:08:22.520 --> 00:08:24.640
We always work to the lowest scope.

s111
00:08:24.640 --> 00:08:32.680
What I mean by that, instead of going through specific scenario that could get confusing, I'm just going to call the access authority for Sven.

s112
00:08:32.680 --> 00:08:39.039
He has access to A, B, C, and D. That could be information he's allowed to see, that could be a tool he's allowed to access.

s113
00:08:39.039 --> 00:08:40.520
But he has that scope.

s114
00:08:40.520 --> 00:08:46.320
Agent 203, notice because we have identities, I can call out this is not agent 409, this is agent 203.

s115
00:08:46.320 --> 00:08:51.280
Agent 203's job is to converse with with Sven about projects.

s116
00:08:51.280 --> 00:09:04.200
And agent 203 has a scope of C and D. And then agent 409 has D, E, and F. So, to get So, when agent 203 calls on behalf of Sven delegating a task,

s117
00:09:04.200 --> 00:09:15.680
agent 203 calls to 409, 409 can actually go to D, E, and F to get the job done, but it can only return D back up the chain here to Sven.

s118
00:09:15.680 --> 00:09:21.960
Because only D, Sven, 203, and 209 all have access to.

s119
00:09:21.960 --> 00:09:29.560
So, the the scope of authority never grows beyond what agent 409 has been given.

s120
00:09:29.560 --> 00:09:34.120
It always is constricted by what is that that lineage.

s121
00:09:34.360 --> 00:09:41.440
Never assume alignment, meaning we have inherited uh alignment here with the model.

s122
00:09:41.440 --> 00:09:47.120
For instance, if you are using a large language model built into that as its own alignments, right?

s123
00:09:47.120 --> 00:09:50.280
It's own rules and its own definitions.

s124
00:09:50.280 --> 00:09:56.000
As we know, the hyperscalers, the frontier models have just said they're misaligned.

s125
00:09:56.000 --> 00:09:58.200
We're going to tell everybody they're misaligned.

s126
00:09:58.200 --> 00:10:02.600
So, we know that we don't have control over that and that's locked in from our perspective.

s127
00:10:02.600 --> 00:10:06.040
Also, that team context I was talking about is coming from somewhere else.

s128
00:10:06.040 --> 00:10:10.680
So, we don't have Sven doesn't have the capability to go change that very easily.

s129
00:10:10.680 --> 00:10:15.160
And then there might be outside agents that are accessing agent 409 from other parties.

s130
00:10:15.160 --> 00:10:19.720
As we grow this whole agentic era, we're going to see this more and more.

s131
00:10:19.720 --> 00:10:24.080
An outside agent kind of needs to be more like a contractor coming to work in your building.

s132
00:10:24.080 --> 00:10:28.240
You give it an identity, you give it its authorization, what it's able to access, and so forth.

s133
00:10:28.240 --> 00:10:39.360
But from the perspective of 409, we need to think that that that alignment, that context coming in is not something Sven controls.

s134
00:10:39.520 --> 00:10:43.560
Opposed to Sven can control what is defined for 409.

s135
00:10:43.560 --> 00:10:46.960
And so, we see he sets up a set of rules that are deterministic.

s136
00:10:46.960 --> 00:10:47.920
That's the floor.

s137
00:10:47.920 --> 00:10:53.360
That's kind of like FGA policies and so forth, where you can define specific things.

s138
00:10:53.360 --> 00:10:55.839
But then you have intents that are probabilistic.

s139
00:10:55.839 --> 00:10:58.560
So, he intends for something to get done.

s140
00:10:58.560 --> 00:11:00.160
And how do we measure that?

s141
00:11:00.160 --> 00:11:02.520
Oh, that's nice.

s142
00:11:02.520 --> 00:11:09.880
And then uh so, the inherited alignment, it can change on us without our control even though we can't change it.

s143
00:11:09.880 --> 00:11:13.680
And the defined alignment becomes a contract for 409.

s144
00:11:13.680 --> 00:11:17.080
So, how do we actually enforce that contract?

s145
00:11:17.080 --> 00:11:19.400
Well, that's with the administration stage here.

s146
00:11:19.400 --> 00:11:23.839
So, with the administration is where we start to enforce everything before.

s147
00:11:23.839 --> 00:11:31.680
So, if as an employee, we think we have trust, but also our enterprises or organizations make sure that they verify that trust.

s148
00:11:31.680 --> 00:11:43.040
So, they have things like uh you know, human OS is over there rules that are making sure that our devices have a mobile device management on it that they're checking what we're doing and

s149
00:11:43.040 --> 00:11:44.560
we have people checking in with us.

s150
00:11:44.560 --> 00:11:49.600
We have supervisors that are watching that we're on a aligned to what we're working on.

s151
00:11:49.600 --> 00:11:59.360
So, we need to look at that for agents as well, but one thing I want to call out here is that behaving that's what we talk about agents we say like are they out of scope?

s152
00:11:59.360 --> 00:12:01.800
Are they accessing things they're not supposed to?

s153
00:12:01.800 --> 00:12:05.000
That does not always equal succeeding.

s154
00:12:05.000 --> 00:12:10.520
So, we need to think of behaving we defined in authority and succeeding we defined in the role.

s155
00:12:10.520 --> 00:12:14.800
Are you actually succeeding at what we want you to do?

s156
00:12:14.800 --> 00:12:22.480
So, the first thing we need to do in administration is we need to know if you're doing what you're supposed to do and you're staying within those those guardrails of authority.

s157
00:12:22.480 --> 00:12:25.040
So, we set oversight.

s158
00:12:25.040 --> 00:12:28.000
Are you doing what we expect?

s159
00:12:28.120 --> 00:12:29.880
Now, that's not a single metric.

s160
00:12:29.880 --> 00:12:31.560
It's actually a scorecard.

s161
00:12:31.560 --> 00:12:39.200
So, on the authority side we have those more deterministic rules like are you staying within side the scope to authority?

s162
00:12:39.200 --> 00:12:49.160
Are you you're only approved approved for up to $40,000 invoices to be dealing with something like that very clear rule and defined.

s163
00:12:49.160 --> 00:12:54.360
But then on the role we tend to lean into things that are more that those intense.

s164
00:12:54.360 --> 00:12:58.200
So, is your invoice match accuracy up to the par we want?

s165
00:12:58.200 --> 00:13:02.520
Is your reconciliation time the cycle time within the scope?

s166
00:13:02.520 --> 00:13:10.720
Yes, you can put rules to these, but more and more as we start to define these we start to look at them and say we we need to look at them as a scorecard.

s167
00:13:10.720 --> 00:13:15.280
So, and the weights will come from the definitions in role and authority.

s168
00:13:15.280 --> 00:13:18.760
So, again rules are checked and tense are judged.

s169
00:13:18.760 --> 00:13:28.240
So, rules like I said these types of never approves a payment rights only to the ledger versus the intense are probabilistic and it's like the correct PO was matched a proper vendor

s170
00:13:28.240 --> 00:13:29.280
project match.

s171
00:13:29.280 --> 00:13:33.960
And an intent evaluation should be based on a confidence level.

s172
00:13:33.960 --> 00:13:37.920
So if we look at that, well, who's actually looking at that?

s173
00:13:37.920 --> 00:13:39.560
The agents need micromanagers.

s174
00:13:39.560 --> 00:13:43.800
We as people, sorry, do not like micromanagers, right?

s175
00:13:43.800 --> 00:13:45.800
We don't want that manager looking over our shoulder.

s176
00:13:45.800 --> 00:13:47.680
But agents need that.

s177
00:13:47.680 --> 00:13:51.800
So we have a micromanager looking at what's going on with the model.

s178
00:13:51.800 --> 00:13:57.440
We have a manager looking at what's going on with the context coming into that agent.

s179
00:13:57.440 --> 00:14:00.800
And then we also look at what's the work product coming out.

s180
00:14:00.800 --> 00:14:05.320
And so there's all sorts of things we should be looking at in the oversight constantly.

s181
00:14:05.320 --> 00:14:14.560
And that's why I have these as little guardian agents around here that are actually applying that at machine speed because if you wait for a quarterly review like we do with humans,

s182
00:14:14.560 --> 00:14:21.480
you are about 40,000, 400,000, 4 million decisions too late to have that discussion.

s183
00:14:21.480 --> 00:14:24.840
So, what happens when things start to go awry?

s184
00:14:24.840 --> 00:14:26.600
We need to take action.

s185
00:14:26.600 --> 00:14:30.800
Well, if you think about it, if we do something wrong, the first step isn't you're fired.

s186
00:14:30.800 --> 00:14:32.440
Hopefully it's not.

s187
00:14:32.440 --> 00:14:36.720
So you you're going to you're going to you know your supervisor is going to sit down and say, "Hey, you know what?

s188
00:14:36.720 --> 00:14:37.800
You misunderstood that.

s189
00:14:37.800 --> 00:14:38.880
This is what we want."

s190
00:14:38.880 --> 00:14:40.040
A correction.

s191
00:14:40.040 --> 00:14:45.000
But if you keep doing it, especially if you're out of your authority, we're going to start restricting that authority.

s192
00:14:45.000 --> 00:14:50.280
And then if you keep doing it, we're going to escalate and then you're out of here.

s193
00:14:50.280 --> 00:14:52.720
So we also need recourse.

s194
00:14:52.720 --> 00:14:55.280
The same kind of model for agents.

s195
00:14:55.280 --> 00:15:00.200
What happens if you don't do what we expect?

s196
00:15:00.480 --> 00:15:13.320
So we look at it as a deterministic grade when we're talking about like intents for the probabilistic judgment, we look at like if if we're pretty confident, then we can take a certain action.

s197
00:15:13.320 --> 00:15:15.160
If we're low confidence, we can take another.

s198
00:15:15.160 --> 00:15:19.880
That allows us in the definitions to be to set these steps.

s199
00:15:19.880 --> 00:15:21.320
And And steps allow actions.

s200
00:15:21.320 --> 00:15:32.440
So for instance, if I'm looking at the intense and I've got a confidence score going, I can look at like if I'm 90% pretty confident, then I might just redact

s201
00:15:32.440 --> 00:15:35.200
content if it if it's out of out of alignment.

s202
00:15:35.200 --> 00:15:39.320
If it's less than 50%, I need to escalate to a human in the loop.

s203
00:15:39.320 --> 00:15:41.560
If it's below 20%, block it.

s204
00:15:41.560 --> 00:15:43.360
We're not We're not confident at all.

s205
00:15:43.360 --> 00:15:53.760
So, by putting in those steps, those stops, we can start to turn that probabilistic into something that can have deterministic actions that can be acted on in recourse.

s206
00:15:53.760 --> 00:16:05.960
But, that's opposed to something like you accessed something you're not supposed to access, and so we restrict authorization at the moment, or suspend access if you've done it five times today,

s207
00:16:05.960 --> 00:16:11.040
or suspend the whole operation of the agent if it's been doing this consistently for days.

s208
00:16:11.040 --> 00:16:17.760
So, that we when we look at a zero trust, we start heavier at first, then we lighten with repeated success.

s209
00:16:17.760 --> 00:16:21.080
So, this is not something you just kind of set and then forget right away.

s210
00:16:21.080 --> 00:16:24.160
It's something that you want to be looking at and evolving.

s211
00:16:24.160 --> 00:16:29.080
And again, at the machine speed is where you can do this practically with agents.

s212
00:16:29.080 --> 00:16:33.360
So, we have the police, the rule, we need a cop.

s213
00:16:33.360 --> 00:16:41.440
That's where we've got something like guardian agents say gateways, guardrails, where they're doing restriction prohibit and suspension at machine speed.

s214
00:16:41.440 --> 00:16:48.720
And then Sven's job, what we're good as at as humans, is we come in with our judgment to be able to say, "You know what?

s215
00:16:48.720 --> 00:16:56.040
If I tweak this rule, if I tweak this authority definition in that alignment stage, then I can get a better result.

s216
00:16:56.040 --> 00:16:59.520
I get more success out of the agent, or I get it to behave better."

s217
00:16:59.520 --> 00:17:04.880
So, Sven adjusts those rules and authority as a coach, if you think about it that way.

s218
00:17:04.880 --> 00:17:08.319
So, we have assignment, we need to name every agent.

s219
00:17:08.319 --> 00:17:11.400
We have alignment, we need to bound every action.

s220
00:17:11.400 --> 00:17:16.000
And we have administration, we need to trace every outcome.

s221
00:17:16.120 --> 00:17:17.400
That's agent accountability.

s222
00:17:17.400 --> 00:17:20.640
That's how we reach agent accountability.

s223
00:17:20.720 --> 00:17:25.000
This was built from the history of humans, tailored for agents.

s224
00:17:25.000 --> 00:17:39.560
Like I said, there's a a framework behind this with controls that I welcome you to take a look at that that you can audit in your organization, and then you can apply those technical solutions that I talked about as the methodology to achieve.

s225
00:17:39.560 --> 00:17:43.760
So, you can get started today by giving your agents an identity.

s226
00:17:43.760 --> 00:17:50.520
And one thing I ask when I give this talk is if everybody can do me a favor, give it the worst picture of his life.

s227
00:17:50.520 --> 00:17:51.120
Thank you very much.

s228
00:17:51.120 --> 00:18:03.280
I'm Tim Tyler in and uh feel free to go to agentaccountabilityframework.org or scan the QR code if you want my landing page, and you can check out the uh the agent accountability page or uh grab it if you want more information.

s229
00:18:03.280 --> 00:18:04.160
We're also downstairs.

s230
00:18:04.160 --> 00:18:06.002
Thank you.

s231
00:18:06.002 --> 00:18:08.002
[applause]