On Feb 24, 2026, Anthropic said it uncovered industrial-scale efforts by Chinese Labs (@DeepSeek, Moonshot and @MiniMax-M2.5) to extract capabilities from its Claude models via distillation.
According to Anthropic, the campaigns involved:
16M+ exchanges
~24,000 fraudulent accounts
Proxy networks to bypass regional restrictions
Targeted extraction of reasoning, coding and tool-use abilities
A lot of digital security products focus on detecting threats after something suspicious has already happened.
But many problems begin earlier: approving the wrong permission, downloading from an unofficial source, trusting a familiar-looking page, or accepting a prompt without enough context.
This makes me wonder whether the real issue is a lack of security tools, or a lack of clear information at the moment a user is making a decision.
My view is that most people do not want more alerts or technical dashboards. They want to understand:
I'm building Draftly, an AI writing assistant for content creators and marketing teams. Here's where I am and what I'm exploring: Current Situation
I've got:
- A solid AI writing tool that content creators love
- Strong initial interest from early users
- Positive feedback on core features
- Growing user base but looking to scale faster What I'm currently testing: Cold Outreach: I'm thinking about reaching out to content marketing agencies and in-house teams. Has anyone cracked the code on getting meetings without being "just another tool" in their inbox? Conversion Strategy: My current focus is on:
- Optimizing the trial-to-paid journey
- Reducing friction in the payment process
- Making onboarding more engaging
- Building sticky features users can't live without :) What I need help with:
1. How are you handling demo bookings that actually show up?
2. What's working for you in terms of trial conversions?
3. Any creative approaches to stand out in the crowded AI writing space? Would love to hear from other founders who've successfully scaled their tools, especially in the content creation space. What unexpected tactics/hacks worked for you? Edit: I'm really interested in strategies that worked in the last 6-12 months, given how quickly the market is evolving.
I wrote 40 posts about building k dlo before I wrote a single email to a stranger. And I know exactly why in that order. Posting feels like go to market. Likes, comments, followers, a nice little chart. But nobody ever said no to me, because I never actually asked anyone. I was publishing to people who already showed up. Selling is writing to one specific person who didn't ask for you. Building in public doesn't train that. Not even slightly. Which is why you can have 8,000 followers and three paying customers, and it isn't bad luck.
Every time I find an interesting skill for an agent, I end up opening a GitHub repo with 20 different files, multiple layers of instructions, examples, configs, and documentation.
And honestly, I just want to know one thing:
What can this skill actually do?
The funny thing is that the more agent-ready these skills become, the harder they seem to be for humans to understand.
Before launching, almost every problem felt important. After putting the product in front of real people, it became much easier to see what actually mattered and what I had just been overthinking.
Still very early, but I m starting to feel like launching is less about saying the product is finished and more about finally getting better information.
Did your priorities change after your first users came in?
