New AI-powered apps designed for adults with ADHD aim to externalise planning and reduce mental load, as tests demonstrate their potential to manage complex everyday tasks amidst uncertainty.
Living with adult attention-deficit/hyperactivity disorder often means coping with poor time management, impulsivity and disorganisation, and the condition can affect work, relationships and everyday planning, according to the Mayo Clinic and Cleveland Clinic. For many adults, the problem is not a lack of ideas but the difficulty of turning them into orderly action. That is why the appeal of AI tools for task management is obvious: they promise to take messy notes, conflicting priorities and half-formed plans, then turn them into something usable.
That promise is already being packaged into a small but growing category of ADHD-focused apps. Recallify says it can analyse notes and recordings to extract tasks and priorities, Taski is built around chat-based planning and focus support, Sukima uses voice input to organise tasks around energy and location, and FlowADHD offers brain-dump planning and smaller, more manageable steps. Taken together, these products reflect a clear design idea: externalise executive function rather than forcing the user to hold everything in working memory at once.
Against that backdrop, the Stuff test is neatly chosen. The exercise was deliberately ordinary: a child’s birthday party at home, with a fixed budget, food restrictions, uncertain guest numbers, limited space and a deadline that includes a same-day cake collection. Those are precisely the sort of overlapping details that can overwhelm someone who is already juggling work, parenting and admin. The point was not which chatbot sounded smartest, but which one could separate the important facts from the noise without inventing certainty where none existed.
ChatGPT came out best because it treated uncertainty as a problem to manage, not a gap to paper over. It identified the main bottleneck as headcount, asked the right follow-up questions about parents and siblings, and understood that the likely indoor crowd would exceed the room’s comfortable capacity. It also rejected the quoted pizza option as too expensive once adult food and other essentials were included, and it gave the strongest guidance on allergies and the risk of cross-contamination.
Claude was the sharpest editor. It cut quickly to the issues that mattered most: who was actually coming, whether the space was sufficient, whether the budget held, and whether parents would stay. Its strength was restraint. Rather than inflating the answer with filler, it produced a cleaner planning framework and showed better judgement about what could not yet be known. Its weakness was that it stopped a little short of a full solution.
Gemini and Copilot were less convincing for different reasons. Gemini looked structured, but introduced basic inconsistencies, including a wrong party finish time, which undermined confidence in the rest of the plan. Copilot was the most elaborate, but much of its neatness came from assumptions not supported by the prompt, including guest numbers, adult attendance and available overflow space. That is a familiar risk with AI planning tools: a polished output can still be built on guesses.
The larger lesson is practical. For people who struggle with planning, the best AI assistant is not the one that produces the most confident checklist. It is the one that spots ambiguity, preserves the unanswered questions and still gives enough structure to reduce the mental load. That is where this experiment becomes more than a novelty test. It suggests that AI can genuinely help with the kind of hidden executive burden that adult ADHD can make so exhausting, provided the user keeps the final judgement.
Disclaimer: This content is intended for informational purposes only. Readers are advised to exercise their own judgement, conduct due diligence, or consult a qualified expert before acting on any information provided.
