Interhuman's Goblin: "Yeah, Friday at Five" — Interhuman AI
inter-1ResearchSafety<br>Interhuman's Goblin: "Yeah, Friday at Five"<br>We found a quirk in Inter-1: when a modality goes missing, the model hallucinates to fill the silence. Hunting it down taught us a lot about omni-modal failure.<br>June 22, 2026·12 min read
:first-child]:!mt-0 [&>:last-child]:!mb-0">Some of our early customers noticed a strange quirk in Inter-1. When they fed it videos with the audio missing and asked it to read social signals, it would occasionally do something it had no business doing: quote someone. In silence, with nothing to hear, it would report that the person on screen had said "Yeah, Friday at five."<br>TL;DR: We found a quirk in Inter-1 — when a modality goes missing, the model hallucinates to fill the silence. Hunting it down taught us a lot about omni-modal failure.<br>Authors: Sid Ravi (AI Research), Patrizio Cugia (Data Engineering)
:first-child]:!mt-0 [&>:last-child]:!mb-0">What Inter-1 does<br>Inter-1 is our model for reading human communication from video. You hand it a short clip and it returns a structured judgment: an overall engagement state, a set of social signals (agreement, hesitation, skepticism, confidence, and so on) each with a grounded rationale.<br>Inter-1 is omni-modal. It inspects the video, listens to the audio, and holistically decides what is going on.
Omni-modal models are powerful precisely because they fuse modalities. But they come with their quirks. When one modality goes missing, the model can fall back on whatever it learned during training to paper over the gap. A video with no audio is exactly that kind of gap, and "Yeah, Friday at five" was what fell into it.
:first-child]:!mt-0 [&>:last-child]:!mb-0">The suspects<br>To hunt down the source of these goblins, our team turned over every stone we could find. The instinct, reasonably, was that a phrase this specific had to be somewhere in the pipeline we trained with. We went looking for it everywhere it could plausibly live.<br>Where we lookedSubset scannedWhat we foundTraining corpus: weekday and datetime mentions30,960 recordsTuesday (118) outweighs Friday (73). "Friday at five": 0.Video transcripts4,603 transcripts0 occurrences of the phrase.Prior inference probes against Inter-1~800 generated records0 occurrences.Untouched storage buckets584 objects / 1.7 GiB0 verbatim hits.Exhaustive log and storage phrase sweep194 hitsMost prompt-load or known sources. 20 inference-like.We focused in quickly on the post-training data since this was one of the most likely culprits. The word "Friday" was over-represented in one part of our post-training dataset, with 73 occurrences, and the weekday distribution across the corpus was genuinely lopsided.
Frequency of date/time expressions in the v6 SFT rationale dataset. Tuesday (118) outweighs Friday (73); "Friday at five" and its clock-time components appear zero times.<br>But Friday was not even the most-mentioned day. If the model were simply echoing the most frequent day it had seen, it would be saying Tuesday. And the verbatim sentence "Yeah, Friday at five" appeared exactly zero times in the approximately 30,000-record training subset we inspected.<br>So we started transcribing the videos themselves. Over-representation there compounding with a missing-audio fallback was also plausible. After 4,603 transcripts and zero hits, "Yeah, Friday at five" started becoming a meme in our Slack channels.
:first-child]:!mt-0 [&>:last-child]:!mb-0">The clue that redirected us<br>The thing that finally turned us around was small. There was a direct mention of Friday in one signal's ontology which the model grounds itself on. Content in the context window gets a great deal of the model's attention, far more per token than a frame of silent video. That was the hint. So we started looking deeper into the harness.<br>We regularly auto-optimize our prompts using the GEPA algorithm, and that turned out to be the first hit. Buried in the May 2026 system prompt (internal name: inter-1 System prompt v3) was an example of what a good response looks like. This would sit in the context window on every inference call so the model picks up the expected format in-context:<br>other_1: "So you're saying the deadline is Friday?" subject: "Yeah, Friday at five. We've got the whole team lined up for it."
Over three weeks of traffic since we moved to this prompt version, the phrase 150x'd in the logs we'd collected. To ablate, we also chased the boring but necessary hypothesis with our backend team — that a cache somewhere was pinning a poisoned example into requests — and traced the API path until we had ruled it out. The phrase was never cached and never learned. It had only ever existed in the prompt we wrote.
Occurrences of "Yeah, Friday at five" in production Cloud Logging (May 26 – Jun 18), binned to 3-day windows. 174 of 194 hits are system-prompt loads; 20 are model output. Volume rises from the late-May ship date of the transcript...