| 1 |
intro
|
how to team up with the robots |
| 2 |
my story
|
jackets on chairs |
| 3 |
my story
|
jackets on chairs solved with intentions |
| 4 |
tell the future
|
the power of intentions |
| 5 |
tell the future
|
intentions give agency |
| 6 |
tell the future
|
intentions absorb context |
| 7 |
tell the future
|
intentions enable course correction |
| 8 |
my story
|
jackets on chairs solved with intentions |
| 9 |
HPT balance
|
a high performing team of humans |
| 10 |
HPT balance
|
balancing accountability and autonomy |
| 11 |
HPT balance
|
too much accountability |
| 12 |
HPT balance
|
too much autonomy |
| 13 |
HPT balance
|
a loop of intentions and progress |
| 14 |
HPT balance
|
a loop of accountability and autonomy |
| 15 |
HPT balance
|
a loop of intentions and progress in practice |
| 16 |
HPT balance
|
coordination loops compound |
| 17 |
HPT examples
|
example: the commander's intent |
| 18 |
HPT examples
|
example: newsrooms |
| 19 |
HPT examples
|
example: agile software development |
| 20 |
HPT use coco
|
it's called continuous coordination |
| 21 |
SP agents
|
now what about agents |
| 22 |
intro
|
robots as in AI agents |
| 23 |
SP agents
|
AI Agents game, single player mode |
| 24 |
SP agents
|
everyone has their own single player agents |
| 25 |
SP agents
|
a team of humans with single-player agents |
| 26 |
MP agents
|
AI Agents game, multiplayer mode |
| 27 |
MP agents
|
companies are deploying multiplayer agents |
| 28 |
MP agents
|
too much autonomy, not enough accountability |
| 29 |
MP agents
|
what happens when multiplayer agents fail |
| 30 |
MP agents
|
multiplayer agents failure rate |
| 31 |
MP agents
|
multiplayer agents success rate |
| 32 |
MP agents criteria
|
key ingredients |
| 33 |
MP agents criteria
|
key ingredient - live, focused context |
| 34 |
MP agents criteria
|
key ingredient - humans in the loop |
| 35 |
MP agents criteria
|
key ingredient - self-correction |
| 36 |
fix fails
|
so what do we do about the multiplayer problem |
| 37 |
fix fails
|
multiplayer failure mode - babysitters |
| 38 |
fix fails
|
multiplayer failure modes: chat channels |
| 39 |
fix fails
|
multiplayer failure mode: memory |
| 40 |
fix fails
|
multiplayer failure mode: 3 strikes |
| 41 |
fix fails
|
trust broken |
| 42 |
solution leverage
|
what can we lever? |
| 43 |
solution leverage
|
we know agents share I/O with people |
| 44 |
solution leverage
|
we know agents pattern match well |
| 45 |
solution leverage
|
we know agents operate on loops already |
| 46 |
solution leverage
|
use the loop of intentions and progress |
| 47 |
solution leverage
|
use the loop of autonomy and accountability |
| 48 |
the big idea
|
the big idea: put them in the same loops |
| 49 |
the big idea
|
the team without the robots in the loop |
| 50 |
the big idea
|
the robots join the team in the loop |
| 51 |
the big idea
|
continuous coordination with robots in practice |
| 52 |
benefits
|
benefits |
| 53 |
benefits
|
benefits: live, focused context |
| 54 |
benefits
|
benefits: humans in the loop |
| 55 |
benefits
|
benefits: self-correction |
| 56 |
outcomes
|
outcomes compound: trust, capacity, effectiveness |
| 57 |
confidence
|
back to the jackets and confidence |
| 58 |
confidence
|
confidence in people and robots |
| 59 |
the big idea
|
a high performing team with robots |
| 60 |
CTA
|
continuous coordination with humans and agents |
| 61 |
CTA
|
get started and get involved |