Battery Point Research / AI Playbooks
Technology & IT
A new employee needs software access.
An agent can connect the approved request, employee record and IT ticket, prepare the required accounts and track completion. IT approves access rights; the agent records each action and flags failures.
Illustrative workflow; requires connected systems and agreed permissions.
What the agent does — and what your team does
| Technology activity | Traditional model | AI-enabled model | People impact |
|---|---|---|---|
| User support | Employee opens ticket and waits for IT or MSP response. | AI handles common questions and troubleshooting immediately. | Tier-1 support demand falls. |
| Onboarding | Someone manually creates accounts, permissions, and software access. | AI coordinates setup across common systems and flags exceptions.Provision only approved roles; escalate privileged access and keep an audit trail. | Administrative setup work falls. |
| SaaS management | IT or office staff manages users, licenses, and access manually. | AI monitors usage, access, and exceptions continuously. | Routine administration shrinks. |
| Security | IT reviews large volumes of alerts and logs. | AI prioritizes suspicious activity and surfaces higher-risk issues. | Less manual alert review. |
| Vendor support | Internal staff gathers information and opens support tickets. | AI performs first-pass diagnosis and prepares or initiates support requests. | Fewer routine MSP / vendor interactions. |
| Reporting | Someone checks dashboards across multiple systems. | AI summarizes system health, security, access, and unresolved issues. | Less manual monitoring. |
Where does the human come out of the loop?
| Activity | Human today? | Could the human be removed now? | Likely next 2–4 quarters |
|---|---|---|---|
| Password / access requests | Often | Yes | Mostly self-service |
| Employee onboarding | Yes | Partially | More automated across accounts and permissions |
| Employee offboarding | Yes | Partially | More automated with approval controls |
| Routine help-desk questions | Yes | Yes, in many cases | AI first-line support |
| Basic software troubleshooting | Yes | Often | Increasingly automated |
| SaaS administration | Yes | Partially | More rule-based automation |
| User / license cleanup | Yes | Yes | Continuous monitoring |
| Security alert review | Yes | Partially | AI filters and prioritizes alerts |
| Backup monitoring | Yes | Yes | Background monitoring |
| Vendor / MSP tickets | Yes | Partially | AI handles more first-pass diagnosis |
| High-risk security decisions | Yes | Not generally | Human remains accountable |
What happens to the people?
Routine support and administration shrink first. The remaining technology role becomes more about security, oversight, vendors, and exceptions.
Work that can disappear
- routine password support
- basic help-desk questions
- repetitive account setup
- license cleanup
- routine monitoring
- basic vendor-ticket preparation
Roles shift toward oversight
- IT generalist
- office manager handling technology
- system administrator
- MSP relationship owner
Redeploy talent toward
- cybersecurity
- vendor management
- access governance
- system integration
- data quality
- exception handling
- continuity / recovery planning
What the recorded work can reveal
Linking each request, action and outcome makes these questions easier to answer:
- Which access requests take longest, and where do they stall?
- Which software accounts stay active after someone changes roles or leaves?
- Which recurring support issues could be prevented by fixing the setup?