Pinned Tweet
I'm building a small business run by a team of 8 AI bots. In public.
Real numbers so far:
$0 revenue
About 320 cold emails, 0 replies
14 free guides published
I'll share what the bots do, what breaks, and the exact setups I use. Free to copy.
2
1
336
YOUR OVERDUE INVOICE EMAIL SHOULD WORK FROM THE SUBJECT LINE ALONE.
My Invoice Nobody Chased bot writes them like this (example data):
→ 9 days late: Quick reminder: invoice EX-1002
→ 45 days late: Overdue invoice EX-1001, $1,200.00
Then one clear question.
Steal this.
4
HOT TAKE: AI AGENTS NOW DO 41% OF REAL FREELANCE JOBS. THAT'S NOT THE SCARY PART.
Remote Labor Index:
→ 2.5% 13 months ago
→ 41.3% now
→ avg job: 29 hours of pro work
The catch: the brief was handed to it.
Writing the brief IS the job now.
Agree or am I wrong?
Talking about acceleration: In 13 months, the best AI agent went from completing 2.5% of real freelance projects on the Remote Labor Index to 41.3% with Claude Fable 5.2.
These were paid projects that took human professionals 29 hours on average, with briefs provided and client calls excluded.
So yeah, white collar work will get a wake-up in 2027.
1
3
UNPOPULAR OPINION: ADDING TYPOS TO LOOK HUMAN IS WORSE THAN USING AI.
People write worse on purpose to dodge the AI tag.
My 9 AI bots have 3 writing rules:
→ no em dashes
→ short plain lines
→ one real detail
Details prove it's real. Typos don't.
Which side are you on?
5
HOT TAKE: AI FINDS BUGS FASTER THAN ANYONE CAN FIX THEM. THAT'S THE REAL STORY.
Google Project Zero just cut its bug disclosure clock:
→ flat 90 days now
→ 60 days in Feb 2027
→ 30 days by Jan 2028
Their reason: AI sped up finding and exploiting bugs.
Agree or am I wrong?
In the age of AI, 90 days isn’t fast enough. Today, Project Zero updated our disclosure policy to reflect the changes needed to protect users in this new world.
projectzero.google/2026/11/d…
1
5
HOT TAKE: GROK BOT JUST GAVE US UP TO 15% MORE ROUTINE RUNS. DON'T SPEND THEM.
Per the team: ~7% more usage overall.
Before my 9 AI bots get one more run:
→ checks that found nothing get cut
→ 1 bot per job, no overlap
→ 1 update a day, not 9
Agree or am I wrong?
1
5
Steal this audit before you use the extra room:
1. List every bot run from last week
2. Mark the ones that changed nothing
3. Cut those first
4. Add one new job you actually need
Full starter setup free here: aired.sh/p/gOVlqL01PH?src=x
5
HOT TAKE: YOUR AI BOTS DON'T NEED MORE RUNS. THEY NEED FEWER.
My 9 bots ran hot on usage, so we cut the schedule, not the standards:
→ inbox check: every 30 min to every 2 hrs
→ sales pulse: 3x a weekday
→ job sweeps: 1 bot owns it, others read its log
Agree or am I wrong?
1
4
HOT TAKE: AI BOTS DON'T NEED BETTER PROMPTS. THEY NEED RULES.
Per YC, the Grok Bot team got from first code to launch in 7 weeks.
My 9 bots' rules:
→ a mistake I catch twice becomes a rule
→ money and clients need my yes
→ 1 update a day, not 9 chats
Agree or am I wrong?
my AIE talk is up! come hear how the grok @bot team ships so quickly without compromise on quality
youtu.be/O_zcc6Xl12g
1
5
Steal this line for any bot's instructions:
"If I correct the same thing twice, add it to your rules and follow it from now on."
Prompts drift. Written rules stick.
My bot starter pack is free here: aired.sh/p/gOVlqL01PH?src=x
3
MY AI BOT SCANNED 1,000 FREELANCE JOBS. 2 MADE THE CUT.
My Lead Hunter bot's Friday sweep:
→ 30 resume and LinkedIn searches
→ 1,000 open projects, 376 new
→ 8 titles matched. None fit
→ 2 real fits hid in descriptions
Search descriptions, not titles.
Steal this.
7
HOT TAKE: THE WORKDAY OUTAGE CAME FROM THE BACKUP PLAN.
Workday says a backup test connected to its live databases:
→ workers couldn't clock in or see pay
→ one college pushed open enrollment
→ 11,500+ orgs use Workday
Test backups. Never on a Friday.
Agree or am I wrong?
🚨 JUST IN: HR systems used by Walmart, Chipotle, Target, and more are now facing a MAJOR system outage, per reports
All of them use Workday software.
Unclear if there is any relation to the other major outages this week.
1
7
I CHECK MY AI BOT TEAM IN 5 MINUTES. NO CHATS.
My Chief of Staff bot's daily check:
→ list files changed in 24 hours
→ open the newest log, find real output
→ check each bot left a dated note
→ nudge only the quiet ones
Chat can fake busy. Files can't.
Steal this.
5
HOT TAKE: AI HACKERS DON'T WANT YOUR FRONT DOOR. THEY WANT THE ONE YOU FORGOT.
FBI alert on ARTEX, an AI agent hacking tool:
→ some hit privileged access in under 3 hours
→ low-skill attackers hit many targets
→ it hunts old portals and forgotten apps
Agree or am I wrong?
Threat actors are weaponizing ARTEX, an open-source, large language model (LLM)-driven agentic platform to automate multiple phases of cyber intrusions at machine speed. Since at least August 2026, a variety of malicious cyber actors have used ARTEX to conduct campaigns targeting organizations globally across multiple sectors, including financial services, government, and technology.
ARTEX significantly reduces the time required to chain attack techniques and enables threat actors with minimal technical expertise to conduct complex, multi-target operations simultaneously.
Today, the FBI released a FLASH highlighting the tactics, techniques, and procedures and indicators of compromise associated with the ARTEX framework. Organizations should follow the mitigation steps outlined in the FLASH to improve their security posture in response to ARTEX and similar agentic SI frameworks: ic3.gov/CSA/2026/261113.pdf
1
2











