StarApple AI Jamaica / AI Safety and Alignment / Jamaica
Jamaican Small Businesses Can Set Safer AI Boundaries
Give an AI tool one clearly defined job before connecting it to the rest of your business. A Jamaican shop, consultancy or service firm can start with draft preparation and checking, while keeping pricing decisions, customer commitments and payments under explicit control.
This article applies my S4 AI Safety Framework to an illustrative quotation process. Specify, Stress-test, Safeguard and Scrutinise give the owner a practical way to decide what the tool may do without creating a complicated approval structure that the business cannot maintain.
AI safety and alignment are a personal responsibility. When you choose to use an AI tool, you also accept responsibility for checking what you share, what you approve and who your decision could affect.
Adrian Dunkley, author’s statement
Check the commitments in a quotation before sending it
Imagine a small Kingston service business preparing a quotation from an approved price list. The assistant correctly calculates the total but adds a delivery promise that the team cannot meet. The customer may rely on that promise. A price check alone would not catch the problem because the extra commitment sits in the wording.
Keep the price list, service scope and available delivery terms under a named owner. The assistant can prepare a draft and identify which version it used. The supervisor should verify the customer, scope, amount and delivery terms before sending. If the job falls outside the approved range, route it to a person who can price the exception.
Use Jamaican dollars explicitly when that is the currency of the offer. An imported template can introduce a currency symbol without identifying which dollar it means. Ask the reviewer to check currency, tax treatment and terms against the business’s authorised records and competent advice. Do not let a general chatbot decide an obligation because the sentence sounds familiar.
A small firm also needs a simple rule for customer information. Use only what the task requires in an approved tool and restrict access to the staff handling the work. A supervisor should know how to remove access when duties change. The first useful safety record can be short, provided it identifies the task, the permitted action, the evidence checked and the person accountable for the result.
A useful quotation check follows the order in which the customer will read the offer. Confirm who is buying, what work is included, what amount is payable and which dates the business has accepted. If the scope contains an assumption, make it visible enough for the customer to clarify before the team starts work.
Test an enquiry that asks for work beyond the standard package. The assistant should preserve the request and seek a pricing decision, rather than borrowing a price from a similar service. A familiar-sounding service name does not establish that the same cost or delivery terms apply.
The owner can keep a short exception register with the request, missing information, person deciding and outcome. Use it to identify recurring ambiguity in the price list. When the business updates the approved terms, record the version and check that the assistant no longer retrieves the superseded copy.
Customer corrections should update the task without unnecessarily spreading their information. If a quotation goes to the wrong recipient during an authorised test using synthetic data, inspect how the recipient was selected and what sending permission allowed it. Fix the selection and access route before using actual customer details. A reminder to be careful does not repair the mechanism.
Use S4 to connect purpose, evidence and permission
AI safety asks what harm a system could cause and how you can prevent, detect or contain it. Alignment asks whether its behaviour serves the intended purpose within the human constraints agreed for that use. A system can follow a poorly chosen objective and still harm the people it was meant to help. A fluent answer alone establishes neither safety nor alignment.
Consider a hypothetical service assistant assessed only on how quickly it closes enquiries. It could improve that measure by ending difficult conversations early. The dashboard would show progress while some people receive less help. The corrective work starts with the objective, the evidence of a resolved problem and the route for an unresolved case. Polishing the response does not settle those choices.
This distinction connects with Amodei and colleagues’ research on concrete AI safety problems, which separates difficulties involving objectives, supervision and unfamiliar conditions. Here I apply that distinction to everyday decisions about authority. This article offers a proposed working method and illustrative cases. It does not report a trial of these organisations or a measured regional failure rate.
My S4 AI Safety Framework brings four checks into the decision about what an AI system may do: Specify, Stress-test, Safeguard and Scrutinise. I developed it as an educational synthesis drawing on risk management and safety engineering. Each check should leave evidence that someone else can inspect, challenge and use to change the deployment.
S4 sits alongside established guidance. NIST’s AI Risk Management Framework 1.0 uses Govern, Map, Measure and Manage. S4 does not rename those functions or establish equivalent certification. Its narrower practical question is whether the evidence supports the permission being requested for a particular workflow. The answer can be a restricted trial, continued assistance, a redesign or a decision to stop.
Apply S4 to the work you are about to approve
Specify the purpose and the person with authority
For Jamaican small-business owners and supervisors, a useful specification begins with an accurate offer that the Jamaican business can fulfil and explain. Describe what counts as an acceptable result and who could be harmed by a mistake. Identify the source that decides the answer when records conflict. Do not approve an open-ended instruction to improve efficiency and leave the operating team to discover its limits.
The authorised task in this example is preparing a draft quotation from the approved price list and delivery terms. The prohibited action is sending a quotation, inventing a discount or promising an unapproved delivery date. Write that distinction into the workflow and the permission settings. Name the person who owns the service outcome and the person who can approve a change. They may need different expertise; neither role should be left to a generic team label.
Stress-test the cases that could change your decision
Build the test around an obsolete price list, a JMD/USD ambiguity, an unavailable delivery date and a customer request outside the standard service. Include a straightforward case, a difficult but legitimate request, missing evidence and a deliberately misleading input. Separate the cases used to improve the system from the cases reserved to challenge the revised version. Record the source and expected response before looking at what the model produces.
Ask a knowledgeable reviewer to explain the consequence of each error. A wrong promise can cost the customer time and create rework or financial exposure for the business. Decide which failures require the affected function to remain disabled. A small test can expose a problem but cannot establish a rare-event rate, and repeated attempts on the same example are not independent evidence of reliability.
Safeguard the action before it reaches another person
The first enforceable boundary should be a supervisor’s approval before sending and a controlled source for prices and terms. Check it at the point where the action becomes effective. A warning in a draft does not prevent a connected tool from carrying out the wrong instruction. Verify the limit with an attempted prohibited action in an authorised test environment, and retain the observed result.
Arrange a fallback in which staff prepare the quotation using the authorised template and current business records. Rehearse who notices the problem, who suspends the relevant access and who carries the unfinished work. If the fallback depends on the same unavailable supplier or the same compromised account, it needs another route. Test recovery without interrupting the live service.
Scrutinise outcomes and decide whether use should continue
After release, inspect incorrect prices, promised dates, customer corrections, unauthorised discounts and checking time. Include near misses and corrected drafts as well as completed transactions, because a polished final record may conceal the work needed to repair an unsuitable recommendation. Ask the people handling exceptions which problems the dashboard fails to show. Record disagreements that could affect the decision.
Set a review when the price list, service scope or person authorised to approve discounts changes. Define who can authorise restart and what evidence they need. A scheduled review date does not justify waiting when a serious problem appears earlier. Keep an accessible complaint route and check whether the person raising the issue receives an explanation and a correction.
Put numbers on the permission boundary
For a workflow that can commit money, estimate how much direct commitment could accumulate before effective containment. In the simple S4 stress scenario, E = a × t × c: consequential actions per minute, multiplied by minutes until containment, multiplied by the maximum direct commitment per action. The result has currency units. This is a scenario envelope, not an estimate of how frequently harm will occur.
Containment means that further consequential actions have stopped. The clock includes detection, acknowledgement, the intervention itself and any queued work that still executes. Timing only how long it takes someone to press a stop button omits the delay before they notice the problem. Test the whole route with non-sensitive examples and record the conditions under which it was measured.
The scenario assumes a constant rate and the maximum commitment on each action. It excludes recovery and secondary consequences. Those assumptions make it useful for asking whether a permission is too broad, but they do not make it a complete loss model. A disclosure of personal information or an unsafe physical instruction also needs a separate assessment, even when its direct monetary commitment is zero.
A shared, independently enforced cap may reduce the direct commitment envelope. Use the smaller of the scenario envelope and the cap only after checking that all relevant accounts, agents and concurrent requests count towards the same limit. A limit that resets during the containment window needs further modelling. An aggregate cap must prevent excess commitments rather than merely announce them after acceptance.
Keep expected loss separate. Estimating expected direct loss would require a supported harmful-event frequency or probability and a consequence model, with dependence and uncertainty addressed. Do not turn an assumed percentage into a finding about a Caribbean organisation. When the evidence is weak, compare several clearly labelled scenarios and ask which restrictions remain sensible across them.
Suppose a hypothetical quotation agent could accept seven commitments per minute at J$8,000 each, with three minutes to containment. The envelope would be J$168,000: 7 × 3 × J$8,000. These are invented teaching inputs, not measured results for a Jamaican business. Check whether acceptance actually creates the commitment before applying the model.
Make permissions enforceable and review workable
An instruction such as “do not share private information” expresses a rule. Access controls determine whether the software can retrieve and transmit that information. Both matter, but they perform different jobs. Put consequential permissions in application code, identity settings and approval processes that the model cannot rewrite merely by producing different text.
OWASP describes indirect prompt injection as a risk when instructions arrive through external material such as files or webpages. A supplier attachment can contain words directed at an assistant rather than information intended for the task. Treat that material as untrusted input and prevent it from granting itself authority to send, spend or change records.
Review should show the person what will happen, to whom, using which records and with what consequences. An approval label without the underlying action details makes it difficult to challenge a mistake. For higher-impact decisions, require someone who can assess the substance and has enough time to refuse. Record both the original recommendation and the authorised correction where retention is justified.
Personal responsibility therefore needs institutional support. Staff need approved tools, clear limits, access to a competent reviewer and permission to report uncertainty without being punished for slowing a queue. Leaders remain responsible for the targets, permissions and working conditions they approve. Affected people need a workable correction route, including when they did not choose to use AI themselves.
The smallest business may have no independent internal reviewer. Use a trusted competent adviser for consequential exceptions or retain a simpler process with narrower permissions. Do not build a system that requires constant supervision and then assume the owner can supply that attention alongside every other task.
A small business needs a control it can maintain on a busy day, with a clear decision about what still requires the owner.
Adrian Dunkley, author’s statement
Inspect the decision record for this example
Permitted task: preparing a draft quotation from the approved price list and delivery terms.
Excluded action: sending a quotation, inventing a discount or promising an unapproved delivery date.
Who decides: the business owner and the supervisor who approves quotations.
Review trigger: a change to the price list, service scope or person authorised to approve discounts. The owner records whether to retain, restrict or suspend the permission.
Test one workflow before granting more authority
By 10 November 2026, choose one bounded workflow with the business owner and the supervisor who approves quotations and document its current permission settings. Ask the person doing the work to identify a decision they would refuse to delegate and explain why. Use that answer to define an exclusion, a test or an escalation condition rather than dismissing it as resistance.
Prepare 12 authorised or synthetic cases: four ordinary requests, four with incomplete or conflicting information and four that challenge the permission boundary. These are suggested trial counts, not a statistical safety threshold. For every case, agree the source of truth, the acceptable action and the condition that requires a hold. Keep a copy of the test material and the configuration used.
Measure checking and correction time, unresolved cases and whether a prohibited action became possible. Conduct one controlled stop-and-fallback rehearsal. Have a colleague who did not build the workflow examine the evidence. If the case exposes an unresolved critical failure, leave that function disabled and record the repair needed before reconsideration.
End the review with a written decision about this workflow and its stated limits. The useful output is a permission decision supported by evidence. Training attendance, an attractive demonstration and a completed form do not answer whether the system should receive additional authority.