{"slug": "12-top-ai-tips-from-high-performers", "title": "12 top AI tips from high performers", "summary": "High performers are separating from AI avoiders in the workplace not by technical comfort but by treating AI as a tool for mechanical work around their judgment, according to a Fast Company roundup of expert tips. Immigration attorney Mahmudul Hasan, founder and CEO of Clarvo, said a petition draft that took more than 20 hours manually now takes around three, freeing 17 hours for evidence strategy and client work. Hasan said firms should pick one repetitive task and run the old and new ways side by side, because \"people support what they can verify.", "body_md": "A clear performance gap is emerging in workplaces between those who actively use [artificial intelligence](https://www.fastcompany.com/section/artificial-intelligence) to boost their output and those who avoid these tools entirely. Drawing on insights from industry experts, this article examines practical strategies that high performers use to integrate AI into their daily work without compromising quality or judgment. These approaches span everything from workflow automation and tool selection to quality control and team knowledge sharing.\n\nThe difference between AI amplifiers and AI avoiders is not comfort with technology. It is what each group thinks the tool is for.\n\nAvoiders assume AI is being pointed at their judgment, so resisting it protects their professional worth. Amplifiers understand it is being pointed at the mechanical work around their judgment, which frees the part they are actually valued for. Same tool, opposite conclusions, and the second group compounds while the first does not.\n\nIn immigration law practice I have seen this play out concretely. A petition draft that takes more than 20 hours manually now takes around three. What the avoider hears in that sentence is that their expertise has been devalued by a factor of seven. What the amplifier hears is that they just got 17 hours back to spend on evidence strategy, on the recommender who needs persuading, on the client conversation nobody else can have.\n\nHere is where it shows up in performance. The amplifier does not just file faster. They start doing work that was previously impossible—running a completed section against the language U.S. Citizenship and Immigration Services actually uses in Requests for Evidence to find weaknesses before filing. Nobody did that manually at 11 p.m. on a deadline. It was theoretically possible and practically never happened. So the gap is not that one attorney is quicker. It is that one is catching problems the other will only discover in a denial.\n\nThat compounds into advancement quickly. Within a year, the amplifier has handled more matters, seen more fact patterns, and built better judgment from the volume. The avoider protected their hours and got less experience for them.\n\nI would add something less comfortable. Some avoidance is well founded. I have watched tools produce petitions that read beautifully and were wrong in ways only a practitioner would catch, and attorneys filed them and took denials. Skepticism in regulated work is not a character flaw. The distinction that matters is between people who are skeptical and testing, and people who are skeptical and abstaining. The first group is amplifying. The second is just waiting.\n\nThe advice I give firms is to stop asking for buy-in on the technology and pick one repetitive task the person already dislikes. Run the old way and the new way side by side, same case, same review at the end. People support what they can verify. Nobody converts from an argument.\n\n[Mahmudul Hasan](https://www.linkedin.com/in/mahmudul-law), founder and CEO, [Clarvo](https://clarvo.us)\n\nThe difference isn’t which tools they bought. It’s whether they were willing to change how decisions actually get made.\n\nAI amplifiers redesign the work. AI avoiders digitize the same broken process and wonder why nothing changed.\n\nI saw this play out directly with a $75 million construction firm we worked with. They had project data spread across more than 15 disconnected systems. Managers were spending the better part of a week pulling reports together before leadership could even see where projects stood. By the time a billing issue surfaced, it was already days old. They weren’t making slow decisions because they lacked information. They were making slow decisions because nobody had redesigned how information moved through the organization.\n\nAn AI avoider organization would have added a dashboard on top of the same fragmented systems and called it transformation. We didn’t do that. We mapped how decisions actually needed to move, built a unified command center around that flow, and compressed their billing cycle from 45 days to seven. Leadership went from a reporting lag of five to seven days, to real-time visibility. The documented annual savings came in at $890,000, not because the AI was impressive, but because the operating model underneath it finally made sense.\n\nThat is the gap between amplifiers and avoiders. One group asks what they can automate. The other asks which decisions need to move faster, who should own them, and what the system needs to look like for that to actually happen. The tools are almost beside the point. The operating model is everything.\n\n[Paul Malott](https://www.linkedin.com/in/paul-malott), CEO, [Automations 24, Inc.](https://automations24.com)\n\nAmplifiers and avoiders split less on who opens the tool and more on who argues with the output instead of accepting it. Plenty of people use AI constantly and are still avoiders in every sense that matters. In my opinion, they just avoid the actual work.\n\nFor example, I watched two account managers at the same ad agency handle a similar deliverable—a competitive landscape summary—within the same week. One ran it through AI, then spent 20 minutes checking two of its claims against the client’s own numbers, and caught a stat that was subtly wrong before it went out. The other treated the AI draft as finished, sent it, and got called out by the client for a number that didn’t match their own dashboard. Both of them used AI. Only one of them did the work that actually builds judgment.\n\nThat’s the divide showing up in who gets pulled into bigger accounts and who stays on the smaller ones: not comfort with the tool, but the habit of catching it being confidently wrong.\n\n[Jeremy Swiller](https://www.linkedin.com/in/jeremyswiller), founder & Chief Innovation Catalyst, [Flux+Form](https://flux-form.com)\n\nThe biggest difference isn’t who uses AI. It’s who is learning to think with it.\n\nAI amplifiers use it to expand what they can explore, question, and create. Initially, they may not work faster. Sometimes we move a little slower because we know AI-generated work needs to be vetted, verified, and refined. The advantage is capability, not speed.\n\nI see this in our own work. AI can get us to a starting point much faster, but that doesn’t necessarily mean we finish faster. We use that gained capacity to go further: ask better questions, explore more possibilities, apply judgment that AI doesn’t have, challenge assumptions, and discover approaches that might not have surfaced otherwise. The time saved isn’t the advantage. It’s what becomes possible because of it.\n\nIt’s a matter of advancement. PwC’s 2025 AI Jobs Barometer found that skills are changing 66% faster in the jobs most exposed to AI, while workers with AI skills command a 56% wage premium. The opportunity isn’t simply becoming more [productive](https://www.fastcompany.com/section/productivity). It’s becoming capable of working at a higher level.\n\nI suspect that’s where the real divide will emerge. AI avoiders may continue doing good work the way they always have. AI amplifiers are changing what they’re capable of doing. Over several years, that difference compounds.\n\n[Sheryle Gillihan](https://www.linkedin.com/in/sgillihan), CEO, [CauseLabs](https://causelabs.com)\n\nThe key differentiator between AI amplifiers and avoiders is how they answer the question, “What’s in it for me?” Amplifiers see direct personal upside. Avoiders do not.\n\nFor many avoiders, the hesitation is rational economics. For others, it is about craft and ownership, the work they built a career on, or the IP they would be handing to a model.\n\nThe result can be a quiet resistance or a skewed review of AI tools to protect their own job. A friend in tech, working in business development, is responsible for winning clients, gathering requirements, and building highly custom systems for them. The company is pushing AI, but he is not on commission and will not be paid more. AI just means more clients and systems to manage for the same paycheck. He is being asked to work harder to widen the company’s margin, with no bonus or promotion on the horizon.\n\nOn the other hand, for amplifiers, AI is all about freedom to get work done and being the ones to make it happen. Inside their companies, good ideas that used to stall for lack of resources are suddenly buildable. The side projects and tedious backlog are suddenly doable. It feels superhuman. For example, I know engineers who chased a bug in their system for a year, one deemed too costly to fix, and Claude found and fixed it in a few hours. That kind of moment converts skeptics.\n\nBut amplifying only pays off when you’re accountable for the output. Everyone has the same AI tools, so raw, unreviewed AI work can lead to the same result, which is why the amplifiers who overtrust it are the first to ship AI slop. Companies wanting to gain the efficiencies of AI without sacrificing quality of work or the morale of their workforce need to communicate that this is not a zero-sum game, show how being more efficient will translate into direct personal upside and job security for each employee, and hold employees accountable for the work that they have AI produce.\n\nIn the meantime, investing in AI skills is worth it, particularly learning to evaluate AI’s output and make it better. Being an avoider is a precarious position. Job postings more and more ask for familiarity with AI, or a sense of how to integrate it well into a workflow. If you get the chance to bring AI into your current job, you will want that story ready for the next one. Avoidance rarely pays off in the long run, though it may hold off the inevitable long enough to figure out your next move.\n\n[Josh Gafni](https://linkedin.com/in/jgafni), cofounder & CEO, [McCoy](https://mccoy.io)\n\nMy expertise is in running technical teams. One example I have from running a core dev team of about 10 people is that a year ago, when AI was not as good as it is now, we had a split into two camps: developers who want to code everything with AI and those who don’t want to touch it because it only creates slop.\n\nThe first group eventually agreed with the second group that AI produces slop and creates a lot of issues. But where the avoiders chose not to use AI, the amplifiers chose to use it and figure out how to compensate for that while still achieving the performance gains it gives.\n\nSpecifically in our work, external quality controls and what we call deterministic quality controls—code quality, architecture, delivery pipelines, testing pipelines, and so on—became far more important. The motto is now that since you can produce an infinite amount of code, if you make bad code, you make a lot of it.\n\nThe whole art now is architecting software and those quality control systems so they don’t allow AI agents to produce the mess you don’t want in your software. A real example of how this gap affects performance is that it creates a lot more of those most wanted and sought-after 10x developers. Junior people now create a lot of code, but it’s bad and it breaks very fast. In our case, a junior person would start extremely well and build something that works, but then in two weeks, nobody knows what it’s doing and nobody can maintain it. It becomes far faster to just code from scratch than to try and maintain their system.\n\nSenior people architect it well in a way that we still know what it’s doing after a couple of months; it’s still testable, controllable, and auditable. Now, a senior person is no longer constrained by the speed of typing and understanding the specifics of the programming language. They are constrained only by their speed of architecting and thinking, and that is a lot faster—like 10 times as fast as they were before.\n\n[Eugene Tartakovskiy](https://www.linkedin.com/in/etartakovsky/), founder, CEO & CTO, [BeRelevant.ai](https://berelevant.ai)\n\nI believe that “AI amplifiers” and “AI avoiders” isn’t simply about who uses the technology and who doesn’t. It’s about the mindset these employees bring to AI.\n\nTrue AI amplifiers are highly motivated and generate their own use cases, refine their prompts, and integrate AI into their daily tasks.\n\nAI avoiders sit at the opposite end and tend to fall into two categories. One group resists AI because they are fearful of losing their job or overwhelmed with the mostly negative narrative in the news media. The other group may have tried using AI but had a poor experience and concluded it’s not worth the effort because it simply doesn’t work or live up to the hype.\n\nThe challenge with avoiders is how they interact with the rest of their team. Some just ignore AI and continue to work the way they always have. They are passive avoiders who can eventually become curious users who may experiment with AI and may over time adopt a few use cases. But the more dangerous group of avoiders are the active ones who fight against AI. They may call out a colleague for their AI usage in a way that accuses them of cheating or questions whether their work deserves any credit. This is dangerous because it can cause champion or even curious users to conceal their AI usage or, worse, stop using the technology altogether to avoid the humiliation. This can hinder the transformation at the team, functional, and organizational level.\n\nThat’s one of the reasons the amplifiers have to play an important role for their team. If all they do is use AI for themselves, their expertise and value remains siloed. But if these amplifiers share their wins, pass along prompts and uses that worked for them, and encourage their teammates to try them for themselves, they can generate the visibility and excitement that can be the difference between a failed pilot or long-term organizational transformation.\n\nTo me, the difference between how these two types of users approach AI is all about the mindset and how the organization presents its impact and use to their employees.\n\n[Kristin Ginn](https://www.linkedin.com/in/ginnkristin), founder, [trnsfrmAItn](https://www.trnsfrmaitn.com)\n\nPeople assume the divide is about who’s good with AI and who isn’t. That’s not really it. The real split is between people who use it to expand their judgment, and people who let their uncertainty about it keep them from engaging at all.\n\nAI amplifiers don’t hand the technology their thinking. They use it to clear lower-value work, surface information faster, and test ideas, which frees up capacity for the parts of the job that still need a human. AI avoiders feel the same uncertainty. They just respond to it by protecting the way work has always been done.\n\nWe felt that tension inside our own organization. Our technology leaders raised real concerns about cybersecurity, privacy, and data protection, the kind of caution you want from the people responsible for protecting an organization. Our operational leaders saw something else: a chance to free up staff time by handing AI the repetitive work, and to connect our systems so decisions could run on real-time data instead of last week’s report.\n\nNeither side backed down, and honestly, neither should have. What actually moved things forward was that both stayed in the conversation instead of retreating to their corner. Our CIO’s team didn’t just say no; they helped build the guardrails, data-protection standards, access parameters, and human review before any AI-generated information touched a real decision. Our operational leaders didn’t just race ahead; they learned to build inside those guardrails instead of around them.\n\nEarly on, it genuinely felt like we might already be behind. That mix of real caution and real opportunity is what let us catch up without cutting corners.\n\nThe people who ended up advancing weren’t necessarily the earliest adopters, and they weren’t always the most technically fluent. They were the ones who stayed in the harder middle ground, asking what could actually improve, what could go wrong, and what still needed a person to decide. If I had to name the real performance gap, that’s it: less about who moves fastest with AI, more about who keeps their judgment attached to it.\n\n[Gearl Loden](https://www.linkedin.com/in/gearl-loden-lodenleadership), leadership consultant/speaker, [Loden Leadership + Consulting](https://www.lodenleadership.com)\n\nThe divide isn’t comfort with AI; it’s more the discipline with the data underneath it. The avoiders skip the tool because it makes them nervous, and I get that. But the people who get a lift out of AI aren’t the ones who trust it the most. They’re the ones who did the boring prep, so the thing could be trusted in the first place.\n\nWhen you point AI at raw, ungoverned data, it hands back answers that are confident and wrong. Clean the data first, tag it, sort out who’s allowed to see what, and the same tool starts giving you answers you can stand behind. The amplifier does that cleanup before turning anything on. The avoider skips straight to the shiny part and never does it.\n\nThe gap shows up fast in healthcare. Two teams got the same analytics capability. The first aimed it at messy data, got answers that were subtly off, and stopped using it within a month. The second spent two weeks on data tagging and access controls before they switched it on. After that, a question that used to take days of back-and-forth report requests took about 30 minutes. Same capability, different prep, and that was the whole difference.\n\n[Mark Sternig](https://www.linkedin.com/in/marksternig), CTO, [Focus](https://focushcs.com)\n\nAI amplifiers build AI into how they already work, making it their new daily default. AI avoiders try an AI tool once, decide the prompting takes more effort than just doing the task, and go back to what they know, whether that’s a spreadsheet or other manual process.\n\nFitting AI into how a team already works isn’t always obvious, and reshaping processes around AI takes significant effort. People who skip AI integration won’t fall behind on output today, but they will eventually, especially as organizations restructure around AI-native workflows.\n\nThis AI divide seems to be driven more by familiarity with technology than specific AI skills. I’ve watched the extremes play out in my company. Some people run a stack of agents to handle small tasks all day, and others tried AI once, then defaulted to old methods, especially in functions like HR and accounting that were never particularly hands-on with technology to begin with. Older generations also got hit with this shift as adults instead of growing up alongside it, and that unfamiliarity compounds in a multigenerational workplace.\n\nIn a recent interview, I witnessed a painful interaction demonstrating this gap. A candidate was asked which AI tool they prefer, whether Claude, ChatGPT, or something else, and the answer was “whatever’s on my iPad.” In the cloud computing industry, we’re full of early adopters, so it’s easy to forget how uneven adoption still is everywhere else.\n\n[Oscar Moncada](https://www.linkedin.com/in/oscarmoncada1), cofounder and CEO, [Stratus10](https://stratus10.com)\n\nI think there are actually three groups here: AI amplifiers, AI avoiders, and what I call AI autopilots.\n\nAI amplifiers are people who are using AI to make themselves more productive and add value to their work. For example, in recruiting, there are so many tasks that take a lot of time. Transcribing is one of them. We use a tool that transcribes our meetings, gives us the notes, and gives us the next steps. And that is so much easier because you don’t have to list everything yourself, and you’re probably going to miss something, right?\n\nIt’s the same thing with creating job descriptions, candidate profiles, scorecards, and all that. You still need to do the human work first, which is understanding the role, what the company needs, what the candidate should look like, and all that. But once you have that, you can use AI to help you do the work around recruiting faster. So basically, you have your main job, which is recruiting, and then you have all these other tasks around recruiting that take up so much time. AI helps you cut down that time and become more productive.\n\nThen you have AI avoiders. These are people who don’t want to use AI. They resist it because maybe they have strong opinions about AI, or they just don’t believe they need it. And I don’t think that’s necessarily always a bad thing.\n\nFor example, if you’re a writer, your main work is writing, brainstorming, and organizing ideas. So maybe you don’t want to use AI for those things because you don’t want to outsource the actual thinking to AI. AI is not that good at writing anyway. So if using AI doesn’t actually make you better at your main job, then I understand why you would avoid it.\n\nThen I think there’s a third group, AI autopilots. These are people who just use AI everywhere because they feel like they have to use AI. They don’t stop and ask, “Does this actually make sense? Is this actually saving me time? Is this actually making my work better?”\n\nAnd it actually does the opposite. It can make your work messier because you’re using AI for something that you could have done better yourself, or you’re using it for something that requires your own judgment and thinking.\n\nSo I think it comes down to knowing where AI makes you better and where it doesn’t. Sometimes using AI is what makes you more productive. Sometimes avoiding it is the better decision. And sometimes, you’re just using it on autopilot because it’s there.\n\n[Friddy Hoegener](https://www.linkedin.com/in/fhoegener), cofounder, [SCOPE Recruiting](https://www.scoperecruiting.com)\n\nThe biggest difference I see between the two is curiosity, and that compounds into influence and advancement in the workplace.\n\nFor example, an AI amplifier tries something, works out what helped, and tells the people around them how to do it. An avoider waits until a tool is mandated, then learns the minimum required.\n\nThere is, of course, an output difference between these two, as AI is helping employees do more with the same or less time. I actually think the bigger effect is how each person is perceived. Sitting out reads as a lack of interest in getting better at your own work, and that impression attaches to everything else you do. It starts to function as a flag when promotions come up, because leaders begin wondering what else you are waiting to be told about.\n\nAvoiders also drift out of the conversation. When the rest of the team is trading what worked and what flopped, they have nothing to contribute.", "url": "https://wpnews.pro/news/12-top-ai-tips-from-high-performers", "canonical_source": "https://www.fastcompany.com/91603166/12-strategies-high-performers-use-to-integrate-ai-into-their-daily-work-without-compromising-quality-yechnology-ai-adoption-quality-strategy", "published_at": "2026-09-21 10:07:00+00:00", "updated_at": "2026-09-21 10:22:45.030871+00:00", "lang": "en", "topics": ["ai-products", "ai-tools", "artificial-intelligence"], "entities": ["Fast Company", "Mahmudul Hasan", "Clarvo", "U.S. Citizenship and Immigration Services"], "alternates": {"html": "https://wpnews.pro/news/12-top-ai-tips-from-high-performers", "markdown": "https://wpnews.pro/news/12-top-ai-tips-from-high-performers.md", "text": "https://wpnews.pro/news/12-top-ai-tips-from-high-performers.txt", "jsonld": "https://wpnews.pro/news/12-top-ai-tips-from-high-performers.jsonld"}}