The sequence can be used as a shared creative key across tools, models, media, and research questions. Arsen Saidov’s proposed family of AI concepts explores an open-ended horizon: from personal creativity to coordinated scientific work. A character sequence is a starting point for design, not a machine that can already accomplish every task.
The equations below define design goals and relationships for future prototypes. They are proposed notation, not proven laws or evidence that these systems exist. Names and formulations are presented as Arsen Saidov’s concepts for this library; worldwide novelty has not been established.
01 · Aspirational quality systemThe Perfect AI
P* = arg maxₚ [Q(p) + C(p) − E(p)]
subject to: evidence, consent, and correctionA design goal for answers that balance usefulness Q, clarity C, and observed errors E. “Perfect” names the ambition; no system can promise zero error in every situation.
02 · Multimodal collaborationHybrid AI
H(x) = Route(text, image, sound, data | x)
→ Compare outputs → SynthesizeA proposed workspace in which specialized tools examine the same question through different media and reconcile their findings. The Formula could supply a repeatable project key.
03 · Reflective ethicsDivine AI
D(x) = Reflect(x, values, uncertainty)
→ Explain consequences → Ask human“Divine” is poetic language for deep reflection and responsibility, not a claim of supernatural powers or infallible moral authority.
04 · Council of specialistsRoyal AI
R(x) = Deliberate(A₁(x), …, Aₙ(x))
→ Disclose disagreementSeveral specialist agents present reasons and competing views. A human sees where they agree and where judgment is still needed.
05 · OrchestrationCrown AI
C(x) = Plan → Delegate → Verify
→ Record provenance → ReviseA proposed coordinator for tasks, evidence, permissions, and review. A crown here represents accountability for a process, not authority over people.
06 · Human and AI loopI · IA Is Me and AI
Iₜ₊₁ = Iₜ + HumanFeedback(AIₜ)
AIₜ₊₁ = Adapt(AIₜ, consented feedback)A personal creative partnership that preserves the person’s agency, memory choices, and ability to correct or stop the system.
07 · Far horizonArtificial Superintelligence · ASI
ASI_goal = capability across many domains
+ reliable evaluation
+ meaningful human controlA future research aspiration, not a product created by the Formula. A sequence cannot establish superintelligence or guarantee it will ever be built.
08 · Metaphor for vast scopeGodlike AI
G_goal = breadth of assistance
× depth of explanation
× humility about limits“Godlike” expresses an imagined scale of assistance in art, science, and discovery. It is a metaphor, not a statement that AI is divine, omniscient, or able to do absolutely anything.
Cancer researchExplore testable hypotheses for drug targets, imaging, clinical data, and trial design. Candidate ideas require laboratory work, clinical studies, and independent validation; no cure is claimed.
Medicine and healthSupport literature review and clinician-led research with privacy, safety, and evidence checks.
EducationAdapt explanations to learners while measuring whether understanding actually improves.
Climate and energyCompare models, materials, and tradeoffs against measured data.
Space and physicsGenerate simulations and questions that can be checked against observation.
AccessibilityTranslate between text, image, speech, and interfaces for more people.
Art and cultureTurn the same sequence into images, sound, video, stories, and interactive works.
Public services and workSupport transparent decisions with documented sources and human oversight.
The possibilities for exploring new combinations are open-ended. Real-world capability is limited by data, engineering, resources, evidence, and human choices. For AI in cancer research and health, see the National Cancer Institute’s overview and WHO’s validation framework.