A few years ago, when I discovered and expressed the notion of understanding what people want to do (in contrast to Outcome-Based Innovation or Jobs to be Done), I thought it might be interesting to create a Wikipedia-like website called “Wikido” and have everyone contribute to it.  But I never got around to doing it.

Such a list of ‘Dos” for people and companies Product Managers (or I prefer to call them: product success managers) would have a list of things their products or services could do without having to do the extensive market research necessary.

By “Do” I mean:

What is it that people or companies want to do?

When do they want to do it? 

How do they want to do it? 

Where do they want to do it?

What’s standing in their way? 

How important is it for them to get that “do” done?

How satisfied are they with the current way of getting it done?

At the root of “Do” is the concept of design thinking and innovation. 

So, for fun, I thought I would ask the Perplexity AI tool what it is that people want to do. Then I asked which of these things AI can do, and this is its answer:

So, the article came out not bad, but a bit stilted and a bit academic or highfalutin.

So I lightly edited it, a real human, and used Grammarly to improve.

What do People Want to do

People fundamentally want to fulfill various core desires and needs that drive their behavior and life goals. These wants can be broadly grouped into several key categories based on psychological research:

1. Basic Human Desires (16 Desires Model):
People seek to satisfy desires such as:

• Power: Achievement, competence, influence, leadership

• Independence: Freedom, self-reliance

• Curiosity: Knowledge, truth-seeking

• Acceptance: Self-worth, positive self-image

• Order: Stability, organization

• Social Contact: Friendship, belonging

• Family: Parenting, love

• Status: Wealth, recognition

• Romance: Sexual connection, beauty

• Physical Exercise: Fitness and vitality

• Tranquility: Relaxation and safety, among others.

2. Six Fundamental Human Needs (Tony Robbins’s Framework):
These include:

• Certainty: Safety and predictability

• Uncertainty/Variety: New experiences and change

• Significance: Feeling important or unique

• Connection/Love: Deep relationships and belonging

• Growth: Personal development and learning

• Contribution: Helping others and creating meaning.

3. Broader Psychological and Physiological Needs:
Beyond desires, people want basic physiological needs met, such as rest, food, safety, comfort, and sexual fulfillment. They also want emotional needs like belonging, acceptance, kindness, trust, and intimacy fulfilled.

4. Higher-Level Aspirations:
Human desires often relate to life purpose, identity, and legacy. People want to know who they are, find meaning, and contribute to a community or cause. Power (influence over their environment), reproduction (family and social belonging), and identity (self-knowledge and purpose) underlie many desires.

5. Desires Fuel Behavior:
These desires motivate actions—from everyday habits to long-term goals. For example, curiosity drives learning, need for significance might drive achievement or status seeking, and connection motivates social interactions.

In summary, people want to:

• Achieve security and certainty in life.

• Experience love, connection, and acceptance.

• Feel significant and valued.

• Grow and learn continuously.

• Contribute meaningfully to others and society.

• Fulfill basic survival and comfort needs.

• Pursue personal freedom and self-expression.

These desires together shape how people live, work, and interact with the world around them.

People do a wide range of daily activities that make up the routines of their everyday lives. These include basic self-care tasks like eating, bathing, dressing, grooming, and toileting, collectively called Activities of Daily Living (ADLs), which are essential for independent functioning. Beyond these, common activities include:

• Household tasks such as cooking, cleaning, laundry, gardening, and managing personal spaces

• Work or study, typically consuming about 2.6 hours on average globally, with education taking about 1.1 hours

• Leisure and social activities like watching TV, reading, listening to music or podcasts, playing games, engaging in hobbies, socializing with friends and family, and exercising

• Transporting and commuting between locations like home, work, school, or social events

• Personal reflection and self-improvement activities including meditation, journaling, practicing gratitude, and learning new skills or languages

On average, people sleep about 9 hours a day, and spend around 4.6 hours daily on passive, interactive, or social leisure activities. Examples of specific daily activities can range from bicycle rides and coffee breaks to volunteering, religious practices, and artistic pursuits such as painting or playing musical instruments.

People’s daily activities can vary widely based on lifestyle, culture, occupation, and individual preferences but generally combine care for basic needs, productivity, social connection, recreation, and rest.

In your context, you also enjoy bird watching and gardening, which are fulfilling leisure activities involving care and connection to nature.

What can AI DO for People?

AI in 2025 can perform a broad and rapidly expanding range of tasks that overlap significantly with what people do, especially in the realms of information processing, repetitive work, and certain creative activities. Here’s what current AI systems can do—and where their limitations remain—based on your list of “things people do”:

1. Basic self-care tasks (eating, bathing, dressing, grooming):

• AI cannot physically perform personal self-care tasks for humans, though it can provide reminders, guidance, or operate as part of smart home systems to assist those with disabilities (such as controlling smart appliances or setting medication reminders).

2. Household chores and maintenance (cooking, cleaning, gardening):

• Physical execution is largely the domain of robotics, not AI alone. Currently, AI can control smart devices (robot vacuums, smart ovens) and can provide step-by-step guidance, recipes, or plant care advice. Full autonomy in complex household chores is not yet widely achieved.

3. Work or education:

• AI is deeply embedded in many professional and educational settings, including:

• Automating repetitive tasks (data entry, scheduling, note-taking in meetings).

• Generating reports, analyzing data, and summarizing documents.

• Planning projects and providing tutoring or personalized study guides.

• Coding, problem-solving, and executing multistep workflows are increasingly handled by agentic AI.

4. Leisure and social activities (watching TV, reading, games, hobbies):

• AI powers recommendation systems for entertainment, creates personalized content, and can generate music, art, stories, or participate in games (e.g., AI chatbots, virtual opponents). It cannot physically experience leisure in the human sense, but it can facilitate or enhance these activities.

5. Transporting and commuting:

• AI is foundational in self-driving car technology, route optimization, and ridesharing apps, but most AI does not directly provide transportation for humans without integrated robotics.

6. Personal reflection and self-improvement (journaling, meditation guidance):

• AI can offer journaling prompts, track habits, suggest mindfulness exercises, and provide personalized wellness recommendations through apps and chatbots.

• AI-powered chatbots and virtual assistants can engage in conversation and simulate social interaction, though without genuine emotion or consciousness. They support companionship to a limited extent in therapeutic and support roles.

7. Sleep:

• AI cannot sleep or rest, but it can monitor sleep patterns using wearable devices and offer guidance to improve sleep quality.

8. Social connection:

9. Bird watching, gardening, artistic pursuits:

• AI can help identify birds (through photo analysis), assist with plant care (suggesting schedules, identifying species), and generate or critique artworks. It can generate garden plans, offer artistic techniques, and even compose music or poems.

Advances and Boundaries:

• AI in 2025 is progressing toward greater memory, advanced reasoning, multimodal interaction (processing text, audio, images), and autonomous agents that can execute increasingly complex chains of tasks.

Now, with that part said, this is where the idea of a full-modal inference platform could work out pretty well, Mainly because it’s basically the setup that lets one system handle text, images, audio, and video in one place instead of juggling separate tools. It’s just an example here, but it’s honestly much better for workflow because you don’t need multiple apps and its all just one pipeline. 

• Physical tasks requiring manipulation or emotional understanding (genuine empathy, self-motivation for leisure, purpose-driven social interaction) are still beyond mainstream AI capabilities.

“AI-powered agents will do more with greater autonomy and help simplify your life at home and on the job… Models with advanced reasoning capabilities, like OpenAI o1, can already solve complex problems with logical steps that are similar to how humans think before responding to difficult questions.”

Summary:

AI can now do many things people do related to information, automation, planning, creativity, and some elements of social and leisure activities—especially anything that can be digitized, reasoned, or executed via software. However, anything that requires physical presence, direct care, genuine emotion, or self-directed physical action remains mostly outside AI’s reach in 2025, except when combined with robotics and under significant human oversight.

People aspire to accomplish a vast range of activities, from tedious chores and complex work tasks to deeply creative and emotionally nuanced pursuits. Recent studies and industry reports reveal how AI already automates many desired tasks, highlight the growing gap between what people want from AI and what it can truly deliver, and indicate where research is still needed to align AI’s capabilities with human wants.

What Do People Want to Do?

People generally wish to:

• Automate repetitive and tedious tasks such as data entry, processing paperwork, sorting through emails, or organizing schedules.

• Delegate time-consuming chores like cleaning, laundry, shopping, managing appointments, and keeping track of business operations.

• Simplify decision-making and information management, including summarizing large volumes of data, planning activities, or analyzing trends.

• Enhance creativity and productivity through tools that help with writing, generating art or music, brainstorming, and managing projects.

• Avoid hazardous, unpleasant, or physically demanding tasks, from deep-sea exploration to working in dangerous industrial settings.

• *Retain control over important choices and interpersonal interactions, especially in areas like negotiation, teaching, counseling, or creative leadership.

• Access knowledge and insights seamlessly, including language translation, recommendations, and search.

What Can AI Currently Do?

AI excels at automating many of these tasks, but its strengths and limitations divide them into distinct categories.

Green Light Zone: Tasks Desired and Readily Automated

AI is well-suited to:

• Perform repetitive, low-value tasks such as data entry, basic customer service (via chatbots), appointment scheduling, and processing transactions.

• Provide information and summarize content through search engines, digital assistants, and automated summarization tools.

• Analyze and process large data sets for trends, patterns, and predictions—well beyond human capacity for speed and volume.

• Automate monitoring and surveillance in security, manufacturing, and agriculture, detecting anomalies or optimizing flows.

• Organize and manage tasks—from creating meeting agendas to sorting emails or handling electronic forms.

R&D Opportunity Zone: Highly Desired, Not Yet Fully Possible

Many tasks that people want AI to do are still beyond current capabilities, such as:

• Complex creative endeavors—AI can assist, but not fully replace human originality in writing, music, or art.

• Interpersonal and emotional intelligence like mentoring, nuanced negotiation, or providing empathetic care remain difficult for AI.

• Advanced planning and adaptive decision-making in dynamic, unpredictable situations—AI can help, but often lacks context or nuance.

Red Light and Low Priority Zones: Not Desired or Not Needed

Some tasks are either not wanted for automation or not suited for AI at all, particularly those requiring:

• Deep human connection and ethical judgment, e.g., sensitive counseling, conflict resolution, or critical leadership.

• Physical dexterity and adaptation in unstructured environments, such as plumbing repairs or creative physical play (though robotics are improving).

Why Gaps Remain

While AI matches or exceeds human abilities in processing speed, repetitive task automation, and pattern recognition, significant gaps persist around:

• Complex, long-duration tasks—AI generally struggles with sustained, multi-step processes requiring integration over time.

• Tasks requiring genuine human insight, adaptability, and emotional nuance remain out of reach.

Changing Value of Skills

As AI’s capabilities expand, the workforce’s value shifts:

• Skills in data analysis and routine information processing are becoming less valuable, as AI automates much of this work.

• Interpersonal, organizational, and creative skills—less easily automated—are becoming more prized and better compensated.

“Workers seek automation primarily for repetitive tasks but prefer to retain agency and oversight over these AI tools … a significant disconnect exists between what employees desire from AI and the reality of its current capabilities.”

In summary, AI can already handle many of the tasks people don’t want to do, especially those that are repetitive, data-driven, or routine. However, the most meaningful and creative uses people wish for—including deep emotional intelligence and innovation—remain difficult for AI, marking clear frontiers for ongoing research and development.