> ## Content Index > Fetch the complete content index at: https://intelligence.artificialrealitymedia.com/llms.txt > Use this file to discover other available public pages before exploring further. # The journalist tool Sam Polstein wants argues back - URL: https://intelligence.artificialrealitymedia.com/journalist-intelligence-tool-pr-agencies-want/ - Published: 2026-09-02T12:04:00.000Z - Updated: 2026-09-02T12:04:00.000Z - Description: Sam Polstein wants a tool that answers a plain question about a named reporter and disagrees with him. A three-week data lag is why nothing qualifies. - Author: Galina Fendikevich - Tags: Public relations, #For AI Companies, #What needs to be built, #Import 2026-09-03 13:12 The tool communications professionals most want does not exist yet: a journalist intelligence system that answers a plain question about a named reporter and then argues with you about the answer. Sam Polstein, who runs communications and content for deep tech startups at HAUS, calls it the thing he would most like engineers to fix, and says a three-week data lag makes what exists today useless. He is not asking for a better ranked list. He has ranked lists. He is asking for a colleague who disagrees. ## What does Polstein want the tool to do? Answer a question about one named journalist, in conversation, and push back on the premise. His framing of the ask, from his interview with Artificial Reality, the AI buyer-evidence podcast hosted by Galina Fendikevich: > "I think there's room for a really powerful tool that I can prompt and just ask a simple question — hey, Ashlee Vance at Core Memory, do you think he'll cover this topic? And it can kind of argue with me as a coworker would." He raises it as an open call rather than a wish: "Maybe I could put this in the 'let's somebody please fix this and get some engineers working on a better tool for this.'" His benchmark is not a model capability. It is a person. "I would love a [Claude](https://claude.ai/?ref=intelligence.artificialrealitymedia.com) that was as good, or close to as good, as a PR person who has a few years of experience." ## What workflow would it replace? An argument on Slack. The judgment Polstein wants automated already exists inside the agency, and it runs as a back-and-forth between people who know the beat: > "That's what we do internally on Slack. I'm constantly going back and forth with the folks on my team of: do you think this person would be good? Well, they just wrote about this, so maybe not. But also they just moved to this outlet and there started this new beat." Look at the structure of that exchange. Recent coverage is a disqualifier, not a match. An outlet move changes the answer. A new beat changes it again. None of those are retrieval problems. They are judgment applied to facts that are days old. ## What would the product have to do? Seven requirements, drawn from what Polstein describes wanting: - Answer a plain question about a named journalist rather than returning a ranked list - Push back and disagree — "I want somebody who's gonna push back and tell me when I'm wrong" - Weight recent coverage heavily; a three-week lag makes it useless - Treat what a journalist has just covered as a disqualifier rather than a match - Track outlet moves and new beats - Combine public data with the agency's own accumulated judgment about each reporter - Reason tangentially — peer companies rather than direct competitors, thematic adjacency rather than keyword overlap Fendikevich summarized the shape as a profile per journalist that aggregates what is publicly available together with what the agency knows to be true from its own experience. Polstein endorsed that description. ## Why have the incumbents not solved this? Because their data lags, and because prompting a general-purpose model is worse. Polstein has used [Muck Rack](https://muckrack.com/?ref=intelligence.artificialrealitymedia.com) and [Cision](https://www.cision.com/?ref=intelligence.artificialrealitymedia.com) throughout his career and describes them as CRMs for exactly this job. His assessment of the data inside them: > "That sort of information, it's still not that easy to come by, still somewhat out of date — although less so than if I was just prompting Claude or [ChatGPT](https://chatgpt.com/?ref=intelligence.artificialrealitymedia.com)." That is a narrow but important verdict. The PR platforms beat a frontier model on journalist data and still fall short of the work. Polstein separately names weak media list building among his criticisms of Claude, the model he otherwise uses every day. ## Why is the argument the product, not the answer? Because the expert user arrives with a hypothesis already formed. What Polstein needs is friction — a system that says *no, they just covered that* — and he is explicit that current models are too effusive to supply it. Being less effusive is, in his words, one of the big reasons he prefers Claude to ChatGPT. This is Artificial Reality's read rather than his: every recommendation product in this category, incumbent or AI-native, is engineered to return a confident ranked answer. Confidence is the default output because it demos well. For a practitioner with a few years of beat knowledge, a confident list is the least useful thing a system can produce, because the list is the part he can already generate himself. ## What should a builder take from this? Two things to test on your roadmap. First, measure your data freshness against the three-week threshold Polstein names, because in this market recency is not a quality attribute, it is the eligibility bar. Second, ship disagreement as a feature: give the system permission to reject a user's premise and cite the coverage that kills it. The commercial detail worth noticing is that Polstein volunteered this ask unprompted and framed it as a call for engineers. That is a lead, not a think piece. Wherever a skilled practitioner already has a hypothesis, the valuable machine is not the one that agrees faster. Whose premise does your product refuse? ## Where this comes from S1E4: He runs PR for deep tech startups — the full interview with Sam Polstein. Listen or watch: [YouTube](https://www.youtube.com/watch?v=Oo5rT9aS2RQ&ref=intelligence.artificialrealitymedia.com), [Spotify](https://open.spotify.com/episode/3QYcgc5ezkSMvH36WJd7qv?ref=intelligence.artificialrealitymedia.com) or [Apple Podcasts](https://podcasts.apple.com/us/podcast/he-runs-pr-for-deep-tech-startups-if-your-agency-isnt/id6780699025?i=1000773200320&ref=intelligence.artificialrealitymedia.com).