ChatGPT work research results are better , more reproducible and accurate , better taste at writing and image generation is better than anything I have tested.
Was very cool to hear about the reasons people love Sol. We're doing the promotion again, except this time for ChatGPT Work:
Tweet what you love about ChatGPT Work, claim $100 in free credits, get more work done.
First 10k get the free tokens: https://t.co/w7QrPBYkrh
GPT-5.6-sol stands out in auditing code among a lot of other things.
5.5 was good but it produced far more false positives , and needed far more audits for meaningful findings.
6/ What works now:
Grok learns purely from engagement sequences.
Post content that makes people reply, watch, or click your profile.
Keep a consistent topic so it can match you to the right audience.
Engage with your own replies, more signals, more reach.
Don't post in burst
5/ What actively kills your reach:
Getting blocked, muted, or reported
Posting 10 times in a row, the algorithm penalizes each post after the first.
Reposting old content , it filters out posts users already saw before scoring even starts.
X now looks at signals like:
- replies
- reposts
- quotes
- clicks
- profile visits
- video views
- photo opens
- shares
- time spent reading
- follows after seeing the post
But it also tracks negative signals:
- not interested
- mute
- block
- report
So in 2026, Does fully AI-written cold outreach really works?
Just tried a tool that sends AI written contextual messages.
Reply rates tanks atleast 4x.
Has anyone tried both ?