What is AI Life Trajectory Predictor?
AI Life Trajectory Predictor is a recursive future inference engine using your selected AI provider. Unlike simple "predict your future" chatbots, it works in rounds: each round, the AI generates 3-5 possible trajectories based on your profile and the path you've chosen so far. You pick the one that resonates, and the AI recursively predicts the next stage —building a tree of branching possibilities that gets deeper and more specific with each round.
After your chosen number of rounds (1-10), the AI delivers a final epilogue that projects 10+ years into the future from your last chosen path, describing the long-term outcome of the trajectory you've been following.
How it works
Step 1: Enter your API key
Choose a provider, model, API key and connection mode. Browser direct is the default. Optional server relay sends your key and text through ToolKnit and its CDN to a supported provider; no application storage or content logging is used.
Step 2: Describe your profile
Write a detailed description of yourself: age, career, goals, constraints, relationships, fears, aspirations. The more specific you are, the more tailored the predictions. Examples:
- "28, software engineer in Shanghai, want to start my own company but afraid of financial risk, single, parents pressure me to settle down"
- "35, marketing manager in London, feeling stuck, considering a career pivot to UX design, married with one child"
- "22, recent graduate in Tokyo, unsure whether to pursue academia or industry, passionate about AI ethics"
Step 3: Choose inference rounds
Select 1-10 rounds using the dropdown. Each round generates 3-5 trajectories. More rounds = deeper branching tree, but also more API usage. 3-5 rounds is a good balance for most profiles.
Step 4: Choose your path
The AI presents 3-5 possible futures, each with:
- Likelihood rating —High, Medium, or Low
- Time span —e.g. "Next 2-3 years" or "5 years out"
- Detailed description —what happens, how it feels, what changes
Pick the path that resonates (click or press A-E on keyboard). The AI then recursively predicts the next stage based on your choice.
Step 5: Read your epilogue
After all rounds complete, the AI generates a final epilogue projecting 10+ years into the future. This is the long-term destination of the path you've been following —a narrative conclusion to your trajectory tree.
Likelihood ratings explained
Each trajectory comes with a likelihood rating:
- High —the AI considers this path highly probable given your profile and current trajectory
- Medium —possible but requires some specific conditions or decisions
- Low —unlikely but not impossible; often represents wild-card or black-swan scenarios
These labels are generated story text, not calibrated estimates or measured probabilities. They do not tell you how likely an event is in real life. Use the branches as prompts for reflection, not evidence for a consequential decision.
Keyboard shortcuts
- A, B, C, D, E —select the corresponding trajectory without clicking
- Reset button —start over with a new profile
Privacy & security
- API key in memory only —never written to localStorage, IndexedDB, or any persistent storage
- Key validated before use —invalid keys rejected immediately
- AI requests send relevant text to your chosen provider. Browser direct sends it from your device; optional server relay passes the key, text and response through ToolKnit and its CDN without application storage or content logging. Provider retention policies and ordinary website analytics remain separate.
- Browser interface —no ToolKnit account is required. AI generation needs your configured provider and internet access.
Frequently asked questions
Is this real prediction?
No. This is AI-generated narrative inference based on the profile you provide. It can surface possibilities you hadn't considered and help you think through different life paths, but it's not a crystal ball. Treat it as a thought experiment, not a life plan.
How much does it cost?
The ToolKnit interface is free, but AI requests may cost money. Rates and quotas depend on your provider and model; input history and generated output affect each request. Check your provider's current pricing and usage. ToolKnit does not supply or guarantee free API credits.
Can I run it multiple times?
Yes. You can run the same profile again and compare 2-3 sets of generated branches. Different outputs are possible, not guaranteed; repeated patterns are not evidence of predictive accuracy.
What if I disagree with the predictions?
That's the point. The trajectories are starting points for reflection, not conclusions. If a prediction feels wrong, ask yourself why — that dissonance is often where the most valuable self-insight lives.