Brief · 6 min read
Compliance in the Age of AI
What the next wave looks like: more automation, more regulation of the automation, and a higher premium on judgment.
Every compliance vendor now has an AI story. Most of them are the same story: software reads the regulation, software flags the risk, software drafts the response. What the demos skip is the part where a wage-and-hour auditor asks a question the model has not seen, and someone in your company has to answer it, on the record.
AI is genuinely good at the volume problem. Twelve thousand pages of employment law is exactly the kind of corpus machines read better than people. Tracking effective dates across jurisdictions, reconciling filings against payroll runs, catching the discrepancy at 2 a.m. — automate all of it. We do.
Machines for detection, humans for judgment — and a clean line of accountability between them.
But the next wave cuts both ways. The same tools that read regulation are starting to be regulated: notice requirements for automated employment decisions, audit trails for algorithmic screening, state rules that treat a model's output as the employer's action. The companies most excited about AI in HR are quietly acquiring a new compliance surface with every deployment.
The durable model is layered: machines for detection, humans for judgment, and a clean line of accountability between them. When the auditor calls, the answer cannot be the model flagged it as low risk. The answer has to be a person who knows why the decision was made and can stand behind it.
Lighter Mondays start with a conversation.
Lighter Mondays start with a conversation.
Start the Conversation
← All notes
Buy the automation. Then make sure someone in the room has sat in front of the board that will ask about it. The premium on that seat is going up, not down.
©2026 SUNNYSIDE · ALL RIGHTS RESERVED
Brief · 6 min read
Compliance in the Age of AI
Compliance in the Age of AI
What the next wave looks like: more automation, more regulation of the automation, and a higher premium on judgment.
Every compliance vendor now has an AI story. Most of them are the same story: software reads the regulation, software flags the risk, software drafts the response. What the demos skip is the part where a wage-and-hour auditor asks a question the model has not seen, and someone in your company has to answer it, on the record.
AI is genuinely good at the volume problem. Twelve thousand pages of employment law is exactly the kind of corpus machines read better than people. Tracking effective dates across jurisdictions, reconciling filings against payroll runs, catching the discrepancy at 2 a.m. — automate all of it. We do.
Machines for detection, humans for judgment — and a clean line of accountability between them.
But the next wave cuts both ways. The same tools that read regulation are starting to be regulated: notice requirements for automated employment decisions, audit trails for algorithmic screening, state rules that treat a model's output as the employer's action. The companies most excited about AI in HR are quietly acquiring a new compliance surface with every deployment.
The durable model is layered: machines for detection, humans for judgment, and a clean line of accountability between them. When the auditor calls, the answer cannot be the model flagged it as low risk. The answer has to be a person who knows why the decision was made and can stand behind it.
Lighter Mondays start with a conversation.
Lighter Mondays start with a conversation.
Start the Conversation
← All notes
Buy the automation. Then make sure someone in the room has sat in front of the board that will ask about it. The premium on that seat is going up, not down.
©2026 SUNNYSIDE · ALL RIGHTS RESERVED