1. Ground
Give AI source material, retrieval, and context that match the task.
The workload library
Twenty AI workloads — and exactly how far each one can go.
Every sample below is classified four ways so you can choose with discipline: is the work deterministic or non-deterministic, when does a human stay in the loop, what can AI safely do today, and how ready is it. The spine underneath all of them is the same — the determinism bridge.
Non-deterministic (the AI half)
Probabilistic. Great for language and judgment support. A 90% first draft is valuable; a wrong answer must be cheap to catch.
- Draft, summarize, classify, qualify
- Answer from approved sources (RAG)
- Recommend next best action
- Detect sentiment, risk, anomalies
The determinism bridge
Deterministic (the systems half)
Reliable. Executes known steps the same way every time. This is where money moves and records are written.
- Validate, approve, calculate
- Write the record, schedule the task
- Enforce permissions and policy
- Log evidence; escalate exceptions
2. Constrain
Allowed actions, forbidden zones, confidence thresholds, required citations.
3. Review
Human approval for money, public posting, deletion, legal, security, irreversible actions.
4. Execute
Deterministic workflows handle API calls, writes, calculations, records.
5. Log
Prompt version, input, output, user, model, cost, evidence, corrections.
6. Improve
Turn misses into better prompts, cleaner data, sharper escalation.
How to read each workload
Four tags, one honest answer
Type
Non-det probabilistic AI
Det rules-based system
Bridged AI + systems together
Human in the loop
HITL: review approve before it acts
HITL: sample spot-check quality
HITL: gate human owns the decision
What AI can do
Do act, with logs
Recommend propose, human chooses
Draft human saves/sends
Gated never alone
Readiness
Ready deploy with controls
Pilot prove it first
Hold not yet / high risk
Featured walkthroughs
Five workloads, in detail
These five are the easiest first wins: high-frequency, reviewable, and measurable. The full sample walkthroughs are here.
The catalog
Fifteen more, ready to map to your business
Same four tags. Read the badges before you buy a tool — they tell you where the human stays, what AI may touch, and whether to deploy or pilot.
Readiness
AI can
Productivity
Meeting notes & actions
Auto minutes with owners and due dates from a transcript.
Non-detHITL: reviewDraftReady
Productivity
Email triage & drafts
Sort, prioritize, and draft replies; you send.
Non-detHITL: reviewDraftReady
Data
CRM enrichment & hygiene
Fill and tidy low-risk fields; flag conflicts.
BridgedHITL: sampleDoPilot
Legal & ops
Contract / doc summary
Summarize terms and risks; never the final word.
Non-detHITL: gateRecommendPilot
Sales
Proposal / RFP drafting
First-pass proposal from your library and the brief.
Non-detHITL: reviewDraftReady
Enablement
Knowledge-base Q&A (RAG)
Answer from your docs with citations; escalate gaps.
Non-detHITL: reviewDoPilot
Support
Ticket classify & route
Right team, first time; audit a sample.
Non-detHITL: sampleDoReady
Customer
Sentiment & churn risk
Flag at-risk accounts; a human saves them.
Non-detHITL: gateRecommendPilot
Finance
Invoice / receipt extraction
Pull line items; approve before posting.
BridgedHITL: reviewDraftPilot
Executive
Weekly operating report
Data extract + variance commentary; leader approves.
BridgedHITL: reviewDraftReady
People
Onboarding / training content
Role-specific guides; an SME reviews.
Non-detHITL: reviewDraftReady
Marketing
Content from voice / notes
Turn a founder's voice into repeatable drafts.
Non-detHITL: reviewDraftReady
People
Resume / candidate screening
Surface a shortlist; never reject alone, bias-checked.
Non-detHITL: gateRecommendHold
Finance & CX
Refund / credit decisions
AI prepares the case; a human approves the money.
Det executeHITL: gateGatedHold
Finance & ops
Forecast / anomaly flagging
Flag outliers for an analyst to confirm.
Non-detHITL: reviewRecommendPilot
The outcomes map
Every workload runs the same path
Trigger to outcome, AI and systems each doing what they're good at, with the human gate exactly where risk lives.
Use it like this: pick one Ready workload where AI can Do or Draft, define the human gate before you build, measure the value, then graduate the next one. Keep everything marked Gated/Hold behind a human until trust and data are earned.
Go deeper
Want more examples?
This is a starter set of twenty. For the full landscape — AI capability domains mapped by category, with maturity signals and the leading tools in each — explore the Skynet Wars AI Capability Index.