Transcript Slide 1

Q: In general, what do you see
as the limitations on current DoD
game-type training aids?
James H. Korris
President & CEO
Creative Technologies Inc.
Q: How could game-type AI
improve?
James H. Korris
President & CEO
Creative Technologies Inc.
Q: How could massively parallel
computing be accomplished on
serious game platforms?
James H. Korris
President & CEO
Creative Technologies Inc.
Q: What role does bandwidth
play?
James H. Korris
President & CEO
Creative Technologies Inc.
Q: COL Tony Krogh of the National Simulation Center
wrote on 2/6/12, “Most impacting is the near complete
loss of SIM/STIM operators in the Mission Training
Complexes and the Battle Simulation Centers in the
schools and CoEs in the next two years.”
What role will AI have in replacing pucksters and
SIM/STIM operators?
D. Ezra Sidran, Ph.D.
Research Scientist
Riverview Artificial Intelligence, LLC
Q: What role will AI have in
combat in the next five years?
Ten years? Twenty years?
D. Ezra Sidran, Ph.D.
Research Scientist
Riverview Artificial Intelligence, LLC
Q: What are the biggest failings
of commercial computer game
AI? What is the solution?
D. Ezra Sidran, Ph.D.
Research Scientist
Riverview Artificial Intelligence, LLC
Q: What areas can AI assist the
DoD?
D. Ezra Sidran, Ph.D.
Research Scientist
Riverview Artificial Intelligence, LLC
Q: What are the biggest
obstacles to greater
implementation of AI at the DoD?
D. Ezra Sidran, Ph.D.
Research Scientist
Riverview Artificial Intelligence, LLC
Q: What are some of the
similarities and differences in the
requirements AI must meet for
DoD games in comparison to
entertainment games?
Michael van Lent, Ph.D.
CEO, President, Chief Scientist
Soar Technology, Inc.
Q: What ramifications do the
differences have for the selection
of AI technologies and the
application of those technologies
in DoD games?
Michael van Lent, Ph.D.
CEO, President, Chief Scientist
Soar Technology, Inc.
Q: What are the biggest success
stories in the application of AI to
DoD problems and DoD games
specifically?
Michael van Lent, Ph.D.
CEO, President, Chief Scientist
Soar Technology, Inc.
Q: If we ask the previous
question about success stories
again in 10 years what do you
think the answer will be?
Michael van Lent, Ph.D.
CEO, President, Chief Scientist
Soar Technology, Inc.
Q: What are the most significant
short-term opportunities for
leveraging virtual environments
for military training and
operations?
Scott Neal Reilly, Ph.D.
Principal Scientist
Charles River Analytics Inc.
Q: What are the current
obstacles, especially related to
artificial intelligence, that need
to be overcome for these
opportunities to be realized?
Scott Neal Reilly, Ph.D.
Principal Scientist
Charles River Analytics Inc.
Q: How can we increase the speed
and decrease the cost of
developing new virtual environments
and virtual agents for new missions
and training needs?
Scott Neal Reilly, Ph.D.
Principal Scientist
Charles River Analytics Inc.
Q: How can we create agents that
represent adversaries that provide
realistic, adaptable behavior for
training, mission rehearsal, and COA
generation?
Scott Neal Reilly, Ph.D.
Principal Scientist
Charles River Analytics Inc.
Q: In training systems, the objective is to
effectively increase the skills of the
individual trainee. How can we use artificial
intelligence techniques to both support the
human trainers and to act as their agents in
customizing the training experience to
maximize training effectiveness?
Scott Neal Reilly, Ph.D.
Principal Scientist
Charles River Analytics Inc.
Q: Technologists tend to proceed
from the assumption that more is
better. But how much AI is really
needed in games?
Robert Franceschini , Ph.D.
VP & Technical Fellow
SAIC
Q: What about the “validity” of AI for
games? Does it makes sense to release
games without solid understanding of the
correspondence of the AI to the real
world? How long are we willing to wait for
validation data?
Robert Franceschini , Ph.D.
VP & Technical Fellow
SAIC