3
Using this capability, it is now possible to build a set of
tools specifically tailored to contract administration:
1. AI Monitors: AI can automatically check each
email and daily report against the contract and flag
any potential issues to the commercial team in real
time.
2. AI Document Chat: Users can chat in plain
English to find answers or generate summaries,
templates and notifications from their contracts.
3. AI Email Search: Users can query the central
repository of emails from the AI Monitor using
plain English to find supporting evidence.
HOW AI TOOLS WOULD HAVE APPLIED
TO THIS CASE
Each of the tools mentioned above can be mapped to one
of the process gaps identified in the case study. Therefore
any one of them could have theoretically prevented the lost
claim for the contractor if they had been implemented.
Identifying the Claim
The AI Monitor would have read one of the early daily
reports and detected the drop in productivity.
It would have read the comment about poor ground
and immediately notified the Project Manager and central
commercial team that there was a potential latent condition
claim.
The claim would therefore have been identified as soon
as the poor ground conditions were encountered.
Processing the Claim
The Project Manager would then have used the AI
Document Chat to check the process required and would
have realized that they needed to notify the client within 7
days.
They would then have asked the AI Document Chat to
generate an instant draft of the notification email.
They would therefore have followed the correct con-
tractual process even under the extra time pressure caused
by the operational issues.
Defending the Claim
The central commercial team would have used AI Email
Search to find any potential correspondence between the
contractor and client mentioning the poor ground condi-
tions if it existed.
The AI would have been able to detect this in emails
with any variation of wording (e.g., “unstable walls,” “bad
ground” etc.) as it doesn’t require exact text matching like
traditional email search tools.
The AI would also have been searching the central
database of communications and would therefore have
been able to analyze correspondence from people who had
left the business.
This could have led to the detection of an earlier noti-
fication and could therefore have successfully defended the
claim.
IMPACT ON PROJECT OUTCOMES
Any one of the tools could have prevented the issue by
resolving one of the identified process gaps. As such, the
contractor would have avoided the two-month period of
financial loss, and therefore added $400k of profit on this
project.
The early flagging of the latent condition would also
have avoided the pressure that this placed on the relation-
ship between contractor and client. This could have had
further reaching impacts such as improving the chances of
winning renewals or future contracts.
DISCUSSION
People often think of contractual issues as being a zero-sum
game where any gains for one side must come at a cost to
the other.
However, cases like this can prove this to be untrue.
There are many situations where proactive contract admin-
istration can lead to solutions that benefit both parties.
For example, if AI had detected the deteriorating
ground conditions early, the client may have determined
that an alternative access was safer and more cost effective.
They could have changed the mine plan and avoided the
situation altogether.
CONCLUSION
This case study demonstrates that AI tools can provide a
clear improvement in contract management, particularly in
dynamic operational environments like mining.
By flagging potential issues early and automating key
contract management tasks, the system not only reduces
financial risk but also strengthens collaboration between
contractors and clients.
In this case, AI would have been instrumental in rec-
ognising a complex contractual nuance—latent condition
qualification—where human oversight had resulted in sig-
nificant financial loss.
Using this capability, it is now possible to build a set of
tools specifically tailored to contract administration:
1. AI Monitors: AI can automatically check each
email and daily report against the contract and flag
any potential issues to the commercial team in real
time.
2. AI Document Chat: Users can chat in plain
English to find answers or generate summaries,
templates and notifications from their contracts.
3. AI Email Search: Users can query the central
repository of emails from the AI Monitor using
plain English to find supporting evidence.
HOW AI TOOLS WOULD HAVE APPLIED
TO THIS CASE
Each of the tools mentioned above can be mapped to one
of the process gaps identified in the case study. Therefore
any one of them could have theoretically prevented the lost
claim for the contractor if they had been implemented.
Identifying the Claim
The AI Monitor would have read one of the early daily
reports and detected the drop in productivity.
It would have read the comment about poor ground
and immediately notified the Project Manager and central
commercial team that there was a potential latent condition
claim.
The claim would therefore have been identified as soon
as the poor ground conditions were encountered.
Processing the Claim
The Project Manager would then have used the AI
Document Chat to check the process required and would
have realized that they needed to notify the client within 7
days.
They would then have asked the AI Document Chat to
generate an instant draft of the notification email.
They would therefore have followed the correct con-
tractual process even under the extra time pressure caused
by the operational issues.
Defending the Claim
The central commercial team would have used AI Email
Search to find any potential correspondence between the
contractor and client mentioning the poor ground condi-
tions if it existed.
The AI would have been able to detect this in emails
with any variation of wording (e.g., “unstable walls,” “bad
ground” etc.) as it doesn’t require exact text matching like
traditional email search tools.
The AI would also have been searching the central
database of communications and would therefore have
been able to analyze correspondence from people who had
left the business.
This could have led to the detection of an earlier noti-
fication and could therefore have successfully defended the
claim.
IMPACT ON PROJECT OUTCOMES
Any one of the tools could have prevented the issue by
resolving one of the identified process gaps. As such, the
contractor would have avoided the two-month period of
financial loss, and therefore added $400k of profit on this
project.
The early flagging of the latent condition would also
have avoided the pressure that this placed on the relation-
ship between contractor and client. This could have had
further reaching impacts such as improving the chances of
winning renewals or future contracts.
DISCUSSION
People often think of contractual issues as being a zero-sum
game where any gains for one side must come at a cost to
the other.
However, cases like this can prove this to be untrue.
There are many situations where proactive contract admin-
istration can lead to solutions that benefit both parties.
For example, if AI had detected the deteriorating
ground conditions early, the client may have determined
that an alternative access was safer and more cost effective.
They could have changed the mine plan and avoided the
situation altogether.
CONCLUSION
This case study demonstrates that AI tools can provide a
clear improvement in contract management, particularly in
dynamic operational environments like mining.
By flagging potential issues early and automating key
contract management tasks, the system not only reduces
financial risk but also strengthens collaboration between
contractors and clients.
In this case, AI would have been instrumental in rec-
ognising a complex contractual nuance—latent condition
qualification—where human oversight had resulted in sig-
nificant financial loss.