Generative AI tools draft text, summarize documents, and write code. They also introduce risks that every business should understand.

Risk 1: Data leakage

Employees paste confidential information into public AI tools. Depending on the service and plan, that data may be stored, reviewed, or used in ways you didn’t intend.

– Reduce it: approve business-grade tools with clear data terms, block unapproved services where possible, and train staff on what must never be entered

Risk 2: Hallucinations

AI can produce confident, plausible, and wrong answers, including invented facts, citations, and numbers.

– Reduce it: require human review of anything important, verify facts against reliable sources, and never rely on AI alone for legal, medical, or financial decisions

Risk 3: Prompt injection

Hidden instructions in a document, email, or web page can manipulate an AI tool that reads them, especially one that can take actions.

– Reduce it: limit what AI tools can access and do, and require approval for sensitive actions

Risk 4: Over-permissioned access

AI assistants connected to your files and email can surface information that people technically have access to but shouldn’t see.

– Reduce it: clean up permissions before enabling AI features

Risk 5: Intellectual property and copyright

Output may resemble existing content, and ownership rules vary.

– Reduce it: review outputs, avoid pasting others’ protected material, and check terms of use

Risk 6: Bias and fairness

AI can reflect biases in its training data, which matters in hiring, lending, and customer decisions.

– Reduce it: test for bias, keep humans involved in decisions affecting people

Risk 7: Privacy law

Personal information handled by AI brings obligations under Quebec’s Law 25 and other privacy laws, including transparency about automated decisions.

Risk 8: Cost and dependency

Usage-based costs can grow fast, and heavy reliance on a single vendor creates lock-in.

A simple governance starter

1. An AI use policy

2. A list of approved tools

3. Data classification rules

4. Human review requirements

5. Training for all staff

DBR Soft can help you write an AI policy and deploy approved tools securely. Contact us for more information.