A knowledge worker who committed to using AI tools for everything shares what actually works, what's overhyped, and what dangers nobody talks about in a Reddit post that's refreshingly free of both hype and doomerism. Most AI takes are either "this will solve everything" or "this will destroy everything." This is from someone who actually used it daily for six months and is willing to admit both the genuine productivity gains and the subtle ways it's making them worse at their job.The genuinely incredible parts: AI eliminated the blank-page problem entirely. Research synthesis that used to take an hour now takes two minutes with Claude Opus 4.6. Non-coders are building automation scripts and custom dashboards. These aren't hypotheticals — these are real productivity gains from daily use.The massively overhyped parts: "AI will do it for you" is still a lie. Everything requires your judgment and context. AI SEO content is even more dead in 2026 than it was in 2024. Fully autonomous workflows exist only in narrow, controlled cases. The promise of set-it-and-forget-it automation doesn't match reality.But the quietly dangerous parts — this is where it gets interesting. The author admits their first-draft writing has gotten worse because they outsourced that skill. Skill atrophy is real when you stop practicing. They now intentionally write without AI some days to maintain the muscle.Then there's the confidence-without-competence problem. Frontier models give confident-sounding answers to things they don't know. If you're not knowledgeable enough to catch errors, you can build strategies on wrong foundations. This is the most insidious risk: AI making you feel smarter while actually making you less competent.The "good enough" trap resonates. AI output is often 80% there. If you stop at 80%, your work looks like everyone else's. The 20% you add is the differentiation — the thing that makes you valuable. But it's tempting to ship the 80% and move on.The vendor dependency warning is something almost nobody talks about. Workflows deeply integrated with specific AI tools and APIs create risk. Pricing changes, policy shifts, service disruptions — these aren't theoretical. The author's entire productivity stack is built on tools that could change terms tomorrow.The honest summary: "AI tools have made me more productive, creative, and capable than I've ever been. They've also made me lazier in ways I didn't notice until recently."That honesty is rare. The people winning with AI in 2026 aren't the ones using the most tools or running the newest models. They're the ones using AI to amplify genuine skills and judgment — not replace them. The technology is impressive. The question is whether we're using it to get better or just to get by.