Editing Openai/6951bbde-3e94-8004-8806-32d66e8e6e16 (section)

== gpt-5-2 ==
=== User: I'm noticing chatgpt produces a lot of tropes common across many social posts its produced. ===
I'm noticing chatgpt produces a lot of tropes common across many social posts its produced. List the top ten turns of phrase I should look for in your outputs to edit when I post something

=== Assistant: Good instinct. ===
Good instinct. There are very recognizable “AI tells” that show up in social posts. Here are the top ten tropes / turns of phrase I’d actively scan for and either delete or rewrite before posting, plus why they’re giveaways and what to do instead.

==== 1. “Let’s unpack this” ====

Why it’s a tell:
Overused opener that signals “thread incoming.”

Edit move:
Delete entirely or replace with a concrete claim.

: ❌ “Let’s unpack why…”
: ✅ “The real issue is…”

==== 2. “At the end of the day” ====

Why it’s a tell:
Vague summative filler with no content.

Edit move:
Cut it. If you need emphasis, state the conclusion directly.

: ❌ “At the end of the day, X matters most”
: ✅ “X matters most because…”

==== 3. “It’s not about X. It’s about Y.” ====

Why it’s a tell:
Classic rhetorical flip that’s become AI boilerplate.

Edit move:
Collapse into a single sentence or add specificity.

: ❌ “It’s not about speed. It’s about alignment.”
: ✅ “Speed fails without alignment across teams.”

==== 4. “Here’s the thing:” ====

Why it’s a tell:
Throat-clearing phrase that delays the point.

Edit move:
Delete and start with the thing. Always.

==== 5. “In today’s fast-paced / ever-evolving world” ====

Why it’s a tell:
Empty context padding, especially on LinkedIn.

Edit move:
Replace with actual context (industry, role, timeframe).

: ❌ “In today’s fast-paced world…”
: ✅ “In Series B biotech, decisions compound fast…”

==== 6. “This isn’t just about…” ====

Why it’s a tell:
Abstract framing without commitment.

Edit move:
Force yourself to finish the sentence with a concrete noun.

: ❌ “This isn’t just about innovation”
: ✅ “This is about killing bad programs earlier.”

==== 7. “Why this matters” ====

Why it’s a tell:
Meta commentary instead of substance.

Edit move:
Delete the heading and state the consequence directly.

: ❌ “Why this matters: teams burn out”
: ✅ “Teams burn out when…”

==== 8. “Actionable / impactful / meaningful” ====

Why it’s a tell:
Generic adjectives that signal abstraction.

Edit move:
Replace with verbs or outcomes.

: ❌ “Actionable insights”
: ✅ “Decisions that change funding, hiring, or timelines”

==== 9. “Three takeaways” / “Key lessons” ====

Why it’s a tell:
Thread-ified content structure screams AI or consultant.

Edit move:
Either:
* Remove numbering, or
* Make one point sharp enough to stand alone.

==== 10. “If you’re X, you should Y” ====

Why it’s a tell:
Broad advice to a hypothetical audience.

Edit move:
Narrow the audience hard.

: ❌ “If you’re a leader, you should…”
: ✅ “If you run a 5–10 person research team…”

==== Bonus high-confidence AI tells ====

Scan aggressively for these and delete on sight:
* “Nuanced”
* “Leverage”
* “Holistic”
* “Unlock”
* “Navigate”
* “Optimize”
* Excessive em dashes — like this — everywhere

==== A simple editing heuristic (use this every time) ====

Before posting, ask:
# Did I state a specific opinion that could annoy someone smart?
# Did I name a real constraint (time, money, org size, risk)?
# Could a human who disagrees quote me and argue back?

If the answer is “no” to any of those, the post probably still sounds AI-smoothed.

If you want, next time you draft a post, paste it here and I’ll brutally de-AI it without softening your point.