Domain 5: Prepare for the Future with AI
Building the skills and mindset to adapt and lead in an AI-driven world
Introduction
You don’t need to be a computer scientist to shape the future of AI, but you do need to be prepared. In a rapidly changing world, AI literacy is not just about tools. It’s about skills, mindsets, and habits that allow you to grow, adapt, and lead in classrooms, workplaces, and communities shaped by AI.
In this chapter, you’ll learn how to spot emerging AI trends, reflect on your goals and strengths, and begin to shape your career and learning pathways with AI in mind. More importantly, you’ll see that preparing for the future with AI doesn’t mean becoming dependent on it. It means learning how to collaborate with it, understand its limitations, and stay flexible and human-centered.
Understanding What’s Ahead
The only constant in AI is change. Tools you use today, like ChatGPT, Gemini, Claude, or Grammarly, might look completely different in a year. DALL-E is a good example. It was the tool that introduced most people to AI image generation. OpenAI retired the DALL-E feature inside ChatGPT on August 30, 2026, replacing it with newer image models (OpenAI, 2026; Zahn, 2026). But the thinking behind these tools is what will remain essential: how AI analyzes data, supports human decisions, and responds to evolving needs.
Current Trends (As of fall 2025):
- Multimodal AI: combines text, audio, and image inputs and outputs. These tools can write essays, summarize videos, generate presentations, and assist in language translation all within a single interface. Notice that no model name is listed here. Version numbers change from one semester to the next, and the major assistants you already use are multimodal by default. Learn the capability, not the version number.
- Agentic AI (e.g., coding agents like Devin, and computer-use agents that operate a browser or a desktop on your behalf): These systems go beyond one-off prompts. They can plan, execute, and adapt to tasks across time, acting more like an assistant than a tool. Agentic AI can:
- Schedule meetings
- Send emails based on context
- Coordinate across apps
- Debug and write full code projects
- Make iterative decisions with minimal input
That list describes what these systems are marketed as doing. Independent testing tells a more modest story, and knowing the difference is part of AI literacy. A team at Answer.AI spent about a month running Devin, sold as an autonomous AI software engineer, on 20 real tasks. Three finished successfully, 14 failed outright, and three were inconclusive (Husain et al., 2025). Broader measurement points the same direction, and it also shows how fast this moves. In March 2025 the research group METR reported that leading AI agents completed software tasks taking a human professional about an hour with roughly 50 percent reliability (METR, 2025). By January 2026 METR’s updated measurements put that figure at 320 minutes, a little over five hours, for the strongest model it tested (METR, 2026). Both numbers are worth knowing. The first tells you the systems were more limited than the marketing suggested. The second tells you that any specific capability number goes stale quickly, which is why you check the source rather than the textbook. Treat any capability list as a claim until you have seen independent evidence for it, and check the date on the evidence.
AutoGPT is worth knowing as history rather than as a current example. In 2023 it went viral as an autonomous goal-chasing loop that would keep working on its own until it decided a task was finished, and for a while it was the public face of agentic AI. It proved expensive and unreliable in practice, and the project has since become a platform where people assemble agents from visual building blocks instead of turning a loop loose. The flashiest early version of a technology is often not the version that lasts.
The rise of agentic AI could shift how we define productivity and autonomy at work. As these tools develop, your ability to oversee, direct, and critique them becomes even more critical.
- AI in Education: From platforms like Khanmigo to personalized study apps, AI is being used to:
- Customize lessons based on learning patterns
- Provide real-time writing feedback
- Translate or simplify course material for accessibility
- Help faculty create course content and assessments
There is real evidence behind some of this. In a randomized controlled trial in a Harvard physics course, students who worked with a carefully designed AI tutor showed median learning gains more than double those of students in a comparable active-learning class session, and most of them finished in less time (Kestin et al., 2025). Notice what made that result possible. The tutor was built around established teaching principles and studied under controlled conditions. A general-purpose chatbot with no instructional design behind it is not the same thing. However, these systems also raise questions about fairness, data privacy, and the need for human context in grading or support.
- AI in Hiring and HR:
- Video interview platforms like HireVue analyze what candidates say and how they say it, drawing on transcripts and speech patterns
- Resume screeners rank applicants using keyword and skills-matching algorithms
- Onboarding bots manage scheduling, benefits, and FAQs
HireVue is a useful case in how public scrutiny changes products. For years the company also scored applicants’ facial expressions and body language during video interviews. In November 2019 the Electronic Privacy Information Center filed a complaint about the practice with the Federal Trade Commission (EPIC, 2021). HireVue stopped using facial analysis in March 2020 and announced the change publicly in January 2021, saying that visual analysis no longer added meaningful value to its assessments (Maurer, 2021). An independent algorithmic audit of the company was published alongside that January 2021 announcement.
Read the sequence carefully, because it is the interesting part. As of August 2026 the FTC had taken no public action on the complaint. The feature came out anyway, after researchers and advocates spent a year asking hard questions in public. Pressure does not always arrive as regulation.
These tools save time, but if poorly monitored, they can encode bias or reject qualified candidates based on limited input. Researchers at the University of Washington tested language models of the kind used to match resumes to job descriptions and found that they favored White-associated names in 85.1 percent of the comparisons tested, favored female-associated names in only 11.1 percent, and disadvantaged Black male-associated names in up to 100 percent of cases for some occupations (Wilson & Caliskan, 2024).
- AI in Health:
-
- Tools like Ada offer preliminary symptom assessment, which is not the same as a diagnosis
-
- Hospitals use AI to flag abnormal scans, optimize billing, and predict care needs
-
- Startups are experimenting with AI-powered mental health chatbots
While these systems offer speed and scalability, they can over-rely on pattern-matching, misread symptoms, or exclude nuance. A systematic review of digital and online symptom checkers found that the tool’s top suggestion matched the correct diagnosis in only 19 to 37.9 percent of cases, while accuracy at judging how urgently a person needed care ranged from 48.8 to 90.1 percent (Wallace et al., 2022). That review covered symptom checkers built before large language models were widely deployed, so treat the numbers as a baseline rather than a current measurement, and ask whether anyone has published newer figures for the specific tool in front of you. These tools can help you decide whether to contact a clinician. They cannot tell you what you have.
Babylon Health is the cautionary case in this space. It was one of the best-known AI health companies in the world, valued in the billions of dollars when it went public in 2021, and it ran NHS-funded primary care in the United Kingdom under the name GP at Hand. It never demonstrated clinical value at the scale its valuation assumed. A deal to take the company private collapsed, it shut down its United States business, and it entered bankruptcy in 2023. Its United Kingdom operations sold to eMed Healthcare UK for roughly 620,000 dollars, a tiny fraction of what the company had once been worth (Jennings, 2023). The risk in an AI health tool is not only whether the technology works. It is also whether the claims survive evidence and whether the company survives long enough to support the people who depend on it.
- AI and Automation:
- In logistics: AI manages supply chains, routes, and inventories
- In creative industries: AI drafts emails, edits images, scores music, and prototypes video
- In administrative work: AI automates documentation, meeting summaries, and form responses
Across all domains, AI can increase efficiency, but only when applied ethically and evaluated critically.
How to Stay Informed:
- Follow trusted sources (e.g., UNESCO, Digital Promise, AI Now Institute, MIT Technology Review)
- Subscribe to newsletters like TLDR AI, a free weekday summary of AI news, research, and tools
- Follow research and advocacy organizations like the Algorithmic Justice League, the nonprofit founded by Dr. Joy Buolamwini that researches algorithmic harm and campaigns for accountability
- Attend webinars or workshops on ethical and educational AI
- Ask questions about new tools you encounter: Who made it? Who does it benefit? What does it replace?
AI Skills and Literacies for the Future
UNESCO and Digital Promise agree: the future of AI requires more than technical know-how. You’ll need human skills to thrive. UNESCO’s AI competency framework for students lists a human-centred mindset and the ethics of AI as its first two competency areas, placing them ahead of AI techniques and AI system design (Miao & Shiohira, 2024). Digital Promise’s AI literacy framework puts human judgment and centering justice at the middle of the whole model, surrounded by the three modes of understanding, evaluating, and using AI (Mills et al., 2024).
Core Literacies:
- Critical Thinking: Can you evaluate an AI tool’s output? Can you identify missing perspectives or values?
- Collaboration: Can you use AI in a team? Can you distinguish between human and AI contributions?
- Adaptability: Are you able to learn new tools? Do you stay curious when technologies shift?
- Ethical Reasoning: Can you spot bias, misinformation, or harmful outputs?
These literacies will matter more than memorizing how any single tool works. They will help you respond to AI’s changes with resilience.
Building Your Toolkit:
- Be strategic and choose the tool(s) that will best fit the task you need to do or the information you need to find
- Learn prompt engineering to make the most of generative AI. Better input leads to better output.
- Explore data literacy (how data is collected, analyzed, and misused)
- Practice version control, save and reflect on your progress when using AI to write, plan, or create
- Engage in scenario planning: how might AI change your major, your industry, your goals?
- Carefully evaluate the information created by generative AI – is it accurate? Can you verify the sources it provides? Is it relevant to your prompt?
Designing Your Career with AI in Mind
AI is impacting every field, not just tech. Lightcast, a labor market analytics company, examined more than 1.3 billion job postings and found that 51 percent of postings requiring AI skills come from outside information technology and computer science, and that postings asking for AI skills advertised salaries about 28 percent higher, close to 18,000 dollars more per year (Lightcast, 2025). Here are some ways to think about your future:
1. Identify Roles That Use AI
- Marketing uses AI to analyze campaigns
- Healthcare uses AI in diagnostics, patient communication, and scheduling
- Education uses AI for grading, feedback, tutoring, and curriculum planning
- Law, AI helps lawyers quickly review documents and find similar past cases that can guide legal decisions.
AI is also expected to change which jobs exist, not only how existing jobs are done. The World Economic Forum’s Future of Jobs Report 2025 projects that by 2030, technological and economic shifts will create roughly 170 million new jobs while displacing roughly 92 million, for a net gain of about 78 million, and that 59 out of every 100 workers will need training to keep pace (World Economic Forum, 2025). PwC’s 2026 Global AI Jobs Barometer, based on more than a billion job advertisements across 27 countries, found that postings asking for AI skills carried a 62 percent wage premium and grew about eight times faster than the job market as a whole (PwC, 2026). Read figures like these as projections and trends, not guarantees, and ask who produced them and why.
2. Identify Your Strengths
Ask yourself:
- What am I good at that AI can’t do well?
- What human qualities do I bring to a field, for example: empathy, creativity, leadership?
- How can AI support, not replace, those strengths?
3. Explore AI Career Tools
- Use tools like LinkedIn’s Career Explorer, Lightcast, or O*NET to see how your job interests are evolving. Lightcast is the company created when Emsi and Burning Glass merged in 2021 and rebranded in 2022, so older handouts and syllabi may still list the former names. Its Career Explorer companion tools are useful, though some of the underlying data is several years old, so treat it as a starting point.
- Try AI-powered resume builders or mock interview coaches, but always bring your voice and values
- Reflect on how to present AI use ethically: “AI-assisted, human-designed.”
Responsible Innovation: You Have a Voice
Whether you work in policy, art, trades, healthcare, or education, you are part of how AI evolves.
- Ask questions: Why was this tool created? Whose values shaped it?
- Offer feedback: Many platforms invite user suggestions, take them seriously
- Lead by example: Model transparent and ethical AI use in whatever role you’re in
- Join movements: Contribute to ethical tech initiatives, equity-centered design projects, or open-source education efforts
Preparing for the future means being aware of your influence. Your voice matters, especially when decisions about AI affect others.
Bringing It All Together: Why AI Literacy Matters
AI is here to stay. What sets you apart is how you use it. Being an AI-literate student means more than knowing how to prompt ChatGPT. It means approaching each tool with discernment, asking hard questions, and standing firm in your values.
As you prepare for the future with AI, remember:
- Critical Thinking is your strongest defense. Don’t let AI do the thinking for you. Always evaluate outputs: What’s missing? Who’s left out? What assumptions are being made? Is the information accurate?
- Preserving Your Voice matters. In a world full of AI-generated content, your lived experience, identity, and creativity are irreplaceable. Use AI to support your voice, not substitute it.
- Logical Reasoning keeps you grounded. Spotting false analogies, slippery slope arguments, or overgeneralizations helps you avoid being manipulated by AI-generated misinformation. Learn to question how an AI makes a conclusion, and whether it makes sense.
- Literacy is Agency. Being AI literate gives you power,to make informed choices, challenge harmful systems, and innovate responsibly. Without it, you risk being a passive consumer of technology.
So use AI. Experiment with it. Collaborate with it. But never hand over your autonomy, values, or judgment. The future will belong to those who can think critically, communicate authentically, and adapt ethically.
Reflection (Optional): Write about one way you plan to prepare for the future with AI. It can be academic, professional, or personal. Address:
- What excites or worries you about AI?
- What skill or habit do you want to develop?
- What role will AI play in your future, and what role will you play in shaping it?
References
Electronic Privacy Information Center. (2021, January 12). HireVue, facing FTC complaint from EPIC, halts use of facial recognition. https://epic.org/hirevue-facing-ftc-complaint-from-epic-halts-use-of-facial-recognition/
Maurer, R. (2021, February 3). HireVue discontinues facial analysis screening. Society for Human Resource Management. https://www.shrm.org/topics-tools/news/talent-acquisition/hirevue-discontinues-facial-analysis-screening
Husain, H., Flath, I., & Whitaker, J. (2025, January 8). Thoughts on a month with Devin. Answer.AI. https://www.answer.ai/posts/2025-01-08-devin.html
Jennings, K. (2023, September 19). Bankrupt digital health company Babylon sells U.K. assets for just $620,000. Forbes. https://www.forbes.com/sites/katiejennings/2023/09/19/bankrupt-digital-health-company-babylon-sells-uk-assets-for-just-620000/
Kestin, G., Miller, K., Klales, A., Milbourne, T., & Ponti, G. (2025). AI tutoring outperforms in-class active learning: An RCT introducing a novel research-based design in an authentic educational setting. Scientific Reports, 15, 17458. https://doi.org/10.1038/s41598-025-97652-6
Lightcast. (2025, July 23). Beyond the buzz: Developing the AI skills employers actually need. https://lightcast.io/resources/blog/beyond-the-buzz-press-release-2025-07-23
METR. (2025, March 19). Measuring AI ability to complete long software tasks. https://metr.org/blog/2025-03-19-measuring-ai-ability-to-complete-long-tasks/
METR. (2026, January 29). Time horizon 1.1. https://metr.org/blog/2026-1-29-time-horizon-1-1/
Miao, F., & Shiohira, K. (2024). AI competency framework for students. UNESCO. https://unesdoc.unesco.org/ark:/48223/pf0000391105
Mills, K., Ruiz, P., Lee, K., Coenraad, M., Fusco, J., Roschelle, J., & Weisgrau, J. (2024). AI literacy: A framework to understand, evaluate, and use emerging technology. Digital Promise. https://doi.org/10.51388/20.500.12265/218
OpenAI. (2026, July 31). ChatGPT release notes. https://help.openai.com/en/articles/6825453-chatgpt-release-notes
PwC. (2026, June 15). AI reshapes global labour market into two distinct paths, rewarding human skills: PwC 2026 Global AI Jobs Barometer. https://www.pwc.com/gx/en/news-room/press-releases/2026/pwc-2026-ai-jobs-barometer.html
United Nations Educational, Scientific and Cultural Organization. (2021). Recommendation on the ethics of artificial intelligence. https://unesdoc.unesco.org/ark:/48223/pf0000381137
Wallace, W., Chan, C., Chidambaram, S., Hanna, L., Iqbal, F. M., Acharya, A., Normahani, P., Ashrafian, H., Markar, S. R., Sounderajah, V., & Darzi, A. (2022). The diagnostic and triage accuracy of digital and online symptom checker tools: A systematic review. npj Digital Medicine, 5, 118. https://doi.org/10.1038/s41746-022-00667-w
Wilson, K., & Caliskan, A. (2024). Gender, race, and intersectional bias in resume screening via language model retrieval. Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society, 7(1), 1578-1590. https://doi.org/10.1609/aies.v7i1.31748
World Economic Forum. (2025). The future of jobs report 2025. https://www.weforum.org/publications/the-future-of-jobs-report-2025/
Zahn, S. (2026, August 5). DALL-E leaves ChatGPT on August 30: Download your images first. Notebookcheck. https://www.notebookcheck.net/DALL-E-leaves-ChatGPT-on-August-30-download-your-images-first.1360522.0.html
Media Attributions
- Sep 13, 2025, 05_51_45 PM