Open questions in consumer-protection, 2026

Building this in the open means we list what we've decided alongside what we haven't. Here are the open questions we're carrying, what we lean on each, and where we'd genuinely like input.

Should companies be able to file consumer counter-statements?

Today, companies can reply once on the complaint thread. They cannot publish a counter-narrative on a separate surface. We lean against — every complaint already carries a public reply slot, and a parallel "company statement" channel risks becoming a marketing surface. Open to a structured "company position" field on the original complaint thread that's visually distinct from the reply.

How long should complaints stay public after the company resolves them?

Today: indefinitely while public, with a "resolved" flag once the consumer confirms. Some platforms hide resolved complaints after 90 days. We lean against — the public record of "we had a problem and they fixed it" is positive signal for the company. Hiding resolved complaints inflates "no public issues" optics for companies that actually do address things, which is the wrong incentive direction.

Should we display a company's complaint volume?

Today: yes, prominently. There's a fairness argument — bigger companies will have bigger absolute volumes purely from scale. Counter-argument: the Resolution Rate is rate, not volume; volume just contextualises whether the rate is statistically meaningful. We lean toward keeping volume visible, with a "X per 1M customers" ratio if/when we have the customer-count source to support it.

Should consumers be able to update a complaint after filing?

Today: only via withdrawal + re-file. Edits are tempting because facts evolve, but they also create a moving target for the company to respond to. We lean toward a structured "update" comment slot that the consumer can post, with the original body intact — closer to a court filing's "addendum" than a wiki edit.

Should we open the AI moderation transparency to the consumer?

Today: rejected complaints carry the moderator's reason; the model's confidence and reasoning aren't user-facing. Opening that surface is interesting — the consumer could see why their complaint was held — but also a vector for prompt-injection attempts. We lean toward exposing the reason category and the moderator note, not the model's internal scores.

If you have a strong view on any of these, write to us at hello@claimbo.com. The decisions land in the docs and in this blog.