1. Which episode best challenges hidden review costs?

Title: Claims Automation’s Dirty Secret: Review Time
Listen: Publisher’s episode page
Who: Insurtech Leadership; guest Ralph von Grafenstein, identified as DocuSketch founder and CEO.
Published: 30 January 2026. Length: 27 minutes 30 seconds, according to the episode listing.
Evidence basis: Publisher synopsis and episode metadata.

Key takeaway: the episode’s central argument is that automation should be judged after verification work is counted. Its description connects additional checking, disputes and elapsed time with the need to redesign workflows and centralise quality, rather than merely accelerate an existing task. (insurtech.podbean.com)

Desk application: before a claims automation pilot, separate time spent drafting, checking, correcting, requesting missing evidence and approving. Compare the complete process with the current internal team—not just the supplier’s processing time.

Buying question: “If your draft arrives faster, how many reviewer minutes remain before the carrier can accept it?”

Caution: this is a supplier-leader perspective. The reviewed synopsis does not provide an independently audited cost-per-accepted-file comparison, sample construction or material-error denominator.

2. What does a carrier say it is actually automating?

Title: Sound Policy podcast: AI transforms insurance claims
Listen/read: FM’s episode and transcript
Who: Brian Amaral interviews Jeremy Gallant, identified as FM’s senior vice president of claims.
Published: 19 April 2026.
Evidence basis: Full, automatically generated publisher transcript.

Key takeaways: Gallant discusses document filing, intake-data capture, report summarisation and drafting file narratives. He distinguishes implemented fast-track processes from experiments and pilots, and describes human checking of estimates and AI-assisted work. Data quality and adjuster competence are recurring themes. (fm.com)

Desk application: classify every proposed automation function as operational, piloted, experimental or merely planned. Then ask what source material a reviewer can inspect and which decisions remain outside the production service.

Buying question: “Can we inspect the original evidence behind each narrative statement, rather than approving a summary in isolation?”

Caution: this is a commercial-property carrier’s account. Do not transplant its loss categories, results or operating assumptions into a routine homeowners business case. We have not used its reported improvements as market benchmarks.

3. Which episode helps frame in-house versus outsourced work?

Title: Episode #103 – The Economics of Independent Adjusting
Listen: Episode on Amazon Music
Who: Bill Auten; The Art of Adjusting podcast.
Published: 10 September 2026, according to the producer’s catalogue. Length: 51 minutes on the episode platform.
Evidence basis: Producer catalogue and distributed episode description.

Key takeaways: Auten examines temporary capacity, geographic coverage, evidence gathering and the use of staff adjusters’ time. The description also covers utilisation, overhead and assignment economics, while distinguishing outsourcing work from outsourcing claims responsibility. (music.amazon.com)

Desk application: build an in-house versus outsourced property claims estimating comparison at task level: intake, drafting, supporting documents, QA, revisions, escalation and final acceptance. Price retained carrier work explicitly.

Buying question: “Which tasks disappear from our operation, which move elsewhere, and which come back when the file fails review?”

Caution: independent adjusting includes activities beyond file production. This is not a price guide for a managed production service or a source of comparable per-file rates.

4. What can a QA leader teach about repeatable file quality?

Title: Why Some Insurance Adjusters Get More Work Than Others with Daniel Hornaday
Listen: The New Adjuster Podcast episode
Who: Hosts Murphy and Kelsey with Daniel Hornaday, identified as CNC’s director of quality assurance.
Published: 6 July 2026. Length: 45 minutes 50 seconds in the publisher catalogue.
Evidence basis: Publisher description; no complete transcript reviewed.

Key takeaway: the publisher frames this interview around habits, communication and file-quality standards associated with repeat assignments, drawing on Hornaday’s file-review and carrier experience. It is a practitioner-oriented alternative to a product demonstration. (shows.acast.com)

Desk application: use it to start a reviewer-calibration workshop. Ask your QA team to bring anonymised examples of a pass, a minor correction and a material return, then document why each received that disposition.

Buying question: “Can the producer and carrier reviewer apply the same acceptance rubric to the same file?”

Caution: the available description does not enumerate a carrier-ready claim file checklist. Do not attribute a detailed defect taxonomy or an acceptance-rate improvement to this episode without further review.

5. Which discussion is useful for AI accountability?

Title: The Future of Insurance Panel on AI & Claims — Recorded Live at the 2026 PLRB Claims Conference
Listen: Episode and published highlights
Who: Bryan Falchuk with Sadiq Isu of All Talentz, Kevin Meyer of PropertyExpert and Shane Emmons of Swept AI.
Published: 14 April 2026; recorded 24 March 2026.
Evidence basis: Publisher’s detailed episode highlights.

Key takeaways: the panel’s published themes include oversight of both humans and AI, input-data quality, model drift, and caution around autonomous actions. It also distinguishes routine-task assistance from the human interaction needed in difficult claims situations. (future-of-insurance.com)

Desk application: turn those themes into an evidence-to-approval control map: who changes rules, who approves releases, who handles exceptions, and who can stop a faulty workflow?

Buying question: “What happens when the model or carrier guideline changes after we approve the pilot?”

Caution: conference discussion is not legal advice. For regulatory context, the NAIC’s AI resource describes its model bulletin’s governance expectations and the requirement that AI-supported actions comply with applicable insurance law. Have counsel check the requirements relevant to your states and workflow. (content.naic.org)

6. Which conversation helps prepare for catastrophe production pressure?

Title: Preparing for Significant and Severe Catastrophe Impacts: A Conversation with Sedgwick’s Scott Richardson
Watch: YouTube recording · Publisher recap
Who: Kimberly George with Scott Richardson, identified as Sedgwick’s president of property for the Americas.
Publisher date: 17 February 2026.
Evidence basis: Detailed publisher recap and linked video.

Key takeaways: the recap connects preparedness with vendor relationships and discusses automation intended to reduce administrative repetition and support quality reviews. The property-technology discussion begins at 15:50, according to Sedgwick’s chapter notes. (sedgwick.com)

Desk application: stress-test the whole queue. Increase incoming files in your scenario, then separately constrain production, technical review, missing-evidence resolution and exception approval.

Buying question: “If your drafting capacity doubles, what happens to our review queue and the oldest unresolved exceptions?”

Caution: this is a broad CAT discussion from a claims-services supplier. It does not establish a guaranteed production capacity, surge acceptance rate or specific contingency commitment.