Skip to content
Contact
☰
Color scheme: Dark.
Applied Research

Complex Questions. Integrated Intelligence.

Interdisciplinary research, intelligence and experimental development, configured around your challenge.

One lab. One request. One agreed result.

Discuss your project

What can EIL do?

Six examples of the engagements customers can bring to EIL. They are not separately purchasable offers.

  • Understand and reconstruct complex matters

    A project has failed and the sequence is disputed.

    Possible results: Reconstruction with evidence level, hypotheses with status.

    02030411

  • Investigate developments, markets and future scenarios

    A new technology is spreading and the consequences are unclear.

    Possible results: Scenarios with disclosed assumptions, observation indicators.

    011012

  • Study human behavior, cognition and interaction

    A team has worked with an AI assistant for months and nobody knows what has changed.

    Possible results: Longitudinal study, perspective models.

    05060708

  • Investigate emergence and complex system dynamics

    Several agents show joint behavior that nobody programmed.

    Possible results: Investigation report with alternative explanations.

    091213

  • Develop intelligent technologies and AI systems

    A review agent system with clear roles is to be developed, from design to a working system.

    Possible results: Architecture, working system at the agreed maturity level, evaluation report.

    131516

  • Independently review systems, assumptions and decisions

    Claims about the behavior of an AI system are to be tested against counterevidence.

    Possible results: Review report with contradiction list and limits of the finding.

    140402

The 16 capabilities of EIL

These are internal capabilities and not separate products. An engagement usually activates several at once. Prices are set per engagement and not per capability.

The label describes the state of the EIL-owned foundation and not the scope of delivery of an engagement. Research-based: rests on running research. Prototype: rests on an experimental system. Exploratory: methodically connectable, no completed project is claimed.

Research, Intelligence & Strategic AnalysisUnderstand, document and assess situations

Exploratory

Market, Competitive & Strategic Intelligence

Read technological and field developments as processes rather than snapshots.

Market, Competitive & Strategic Intelligence

Where is a field changing, and which developments are signal and which are noise?

Use cases

  • Assess early indications of a technology shift in a field.
  • Check the assumptions of a strategy against the available evidence.

Possible results

Intelligence report that separates observation, explanation and evidence.

Foundation

Emergence research, the observation-explanation-evidence separation, adversarial thinking.

Maturity and limits

Exploratory No market forecast and no investment advice.

Can be combined with

10 Prediction, Scenario & Future Intelligence14 Adversarial Intelligence, AI Safety & Evaluation11 World Models & Causal Reconstruction

Research-based

Project & System Intelligence

Investigate the state, failure patterns and hidden constraints of a running project or system.

Project & System Intelligence

A project or system behaves differently than planned and the cause is not visible.

Use cases

  • Read the software traces of a running system like measurements to find failure patterns and hidden constraints.
  • Reconstruct the decision history of a project from documents and logs.

Possible results

Diagnostic report, reconstruction of state and sequence, list of open assumptions.

Foundation

AI-augmented system diagnostics, reconstruction logic, SELFGRAPH, kernel with provenance and replay.

Maturity and limits

Research-based Results are stated per engagement with evidence status. No promise of effect is made.

Can be combined with

03 Crisis, Failure & Escalation Intelligence15 Research Infrastructure & Intelligence Workbenches13 Multi-Agent Intelligence & AI Engineering

Exploratory

Crisis, Failure & Escalation Intelligence

How small deviations turn into a crisis, and which steps contributed.

Crisis, Failure & Escalation Intelligence

A failure or escalation has occurred and the sequence is unclear or contested.

Use cases

  • Trace a failed development in a project over time and mark turning points.
  • Compare several possible courses of an escalation as hypotheses with a status.

Possible results

Reconstructed course with marked evidence level, competing explanations, open questions.

Foundation

Cognitive reconstruction, DINGIR (competing hypotheses, cooperation with RFI-IRFOS), MIRROR (drift and contradiction).

Maturity and limits

Exploratory Reconstructions are explanations and not documented sequences. No attribution of blame to persons.

Can be combined with

04 Forensic & Evidence Intelligence11 World Models & Causal Reconstruction12 Institutional & Socio-Technical Dynamics

Research-based

Forensic & Evidence Intelligence

Sort and prepare evidence so that its origin and strength remain traceable.

Forensic & Evidence Intelligence

A lot of material, scattered sources and the question of what actually proves something.

Use cases

  • Build an evidence matrix for a disputed claim, including counterevidence.
  • Reconstruct a temporal sequence of events from distributed documents.

Possible results

Evidence matrix, provenance overview, reconstruction model with confidence status.

Foundation

Provenance and replay in the kernel, DAEDALUS (evidence matrix), JANUS (origin of handoffs), reconstruction methodology.

Maturity and limits

Research-based No court-admissible expert work, no diagnosis of persons, no legal advice.

Can be combined with

03 Crisis, Failure & Escalation Intelligence11 World Models & Causal Reconstruction14 Adversarial Intelligence, AI Safety & Evaluation

Human, Behavioral & Cognitive IntelligencePeople, behavior and thinking in interaction

Research-based

Behavioral & Interaction Intelligence

Analyze interaction patterns, decision dynamics and change over longer periods.

Behavioral & Interaction Intelligence

Single interactions show no pattern, only the course makes deviation and tendency visible.

Use cases

  • Longitudinal analysis of usage trajectories of an AI system over months.
  • Examine how team work practices change after a new tool is introduced.

Possible results

Interaction study with trajectory view, pattern description and limits of the finding.

Foundation

LIIE, UIP, EIA, SELFGRAPH and longitudinal interaction analysis.

Maturity and limits

Research-based Describes patterns in interaction and does not diagnose persons. LIIE is in preparation.

Can be combined with

06 Cognitive & Perspective Intelligence07 Human-AI Interaction & Experience Engineering10 Prediction, Scenario & Future Intelligence

Exploratory

Cognitive & Perspective Intelligence

Reconstruct perspectives, assumptions and thinking structures from observable behavior and material.

Cognitive & Perspective Intelligence

Why do those involved decide as they do, and which view of the situation lies behind it?

Use cases

  • Compare competing views in a conflict as perspective models.
  • Trace how a way of thinking changed over the course of a project.

Possible results

Perspective models with evidence reference, comparison of assumptions, open points.

Foundation

Perspective modeling and cognitive reconstruction, introspection and cognitive externalization.

Maturity and limits

Exploratory Reconstructs perspectives as models and makes no statements about inner states of persons.

Can be combined with

05 Behavioral & Interaction Intelligence11 World Models & Causal Reconstruction03 Crisis, Failure & Escalation Intelligence

Research-based

Human-AI Interaction & Experience Engineering

Study and shape human-AI interaction so that its effect can be observed over time.

Human-AI Interaction & Experience Engineering

An AI system is in use and nobody knows how usage and work practices change because of it.

Use cases

  • Long-term study of the co-evolution of users and an AI assistant.
  • Review the interaction design of an AI tool against observed work practice.

Possible results

Study report, observation protocol, interaction design recommendations with evidence status.

Foundation

Human-AI co-evolution, cognitive externalization, LIIE, Observatory as measurement environment (prototype).

Maturity and limits

Research-based The Observatory is a prototype. Effect is investigated and not promised.

Can be combined with

05 Behavioral & Interaction Intelligence13 Multi-Agent Intelligence & AI Engineering14 Adversarial Intelligence, AI Safety & Evaluation

Exploratory

Neuroadaptive & Cognitive Interfaces

How interfaces and adaptive systems influence perception, behavior and cognitive work.

Neuroadaptive & Cognitive Interfaces

An adaptive system is meant to respond to people, and it is open how this changes their work.

Use cases

  • Concept study on adaptive interface strategies and their cognitive effect assumptions.
  • Design an experiment for the evaluation of an adaptive interface.

Possible results

Concept paper, experiment design, evaluation plan.

Foundation

Research space Human-Machine Interfaces & Neuroadaptive Systems, Neuroadaptive Bridge (in preparation).

Maturity and limits

Exploratory No medical product, no diagnostics and no hardware development in the offer. The field is exploratory.

Can be combined with

07 Human-AI Interaction & Experience Engineering06 Cognitive & Perspective Intelligence13 Multi-Agent Intelligence & AI Engineering

Emergence, Prediction & Complex SystemsOrigin, development and dynamics of complex systems

Research-based

Emergence & Complex Systems Research

Investigate when interaction gives rise to new structures or capabilities and under which conditions.

Emergence & Complex Systems Research

A system shows behavior that cannot be explained from its parts, and its origin is open.

Use cases

  • Test whether observed group behavior of agents is more than instruction or chance.
  • Map feedback loops in an organization or a software system.

Possible results

Investigation report with alternative explanations, candidate phenomena and evidence level.

Foundation

Emergence research as the core of the EIL perspective, EIA, IEIA, CEI, space C.

Maturity and limits

Research-based EIL speaks of candidate phenomena until the evidence supports more. Emergence is not assumed.

Can be combined with

13 Multi-Agent Intelligence & AI Engineering10 Prediction, Scenario & Future Intelligence12 Institutional & Socio-Technical Dynamics

Exploratory

Prediction, Scenario & Future Intelligence

Model possible developments as scenarios with disclosed assumptions.

Prediction, Scenario & Future Intelligence

How could a situation develop, and which assumptions carry this assessment?

Use cases

  • Build scenarios for the spread of a technology with explicit assumptions.
  • Run a behavior model of a user group through as a scenario.

Possible results

Scenario set with assumptions and uncertainties, assessment of directions of development, observation indicators.

Foundation

Behavioral intelligence and prediction, state models, Observatory (prototype).

Maturity and limits

Exploratory Scenarios are not predictions with a hit rate.

Can be combined with

01 Market, Competitive & Strategic Intelligence05 Behavioral & Interaction Intelligence11 World Models & Causal Reconstruction

Research-based

World Models & Causal Reconstruction

Reconstruct a hidden state and possible causal structure from incomplete evidence.

World Models & Causal Reconstruction

Important parts of a complex situation are not observable and have to be inferred.

Use cases

  • Map an information situation with many actors as a world model with confidence status.
  • Build a knowledge graph from documents with provenance per edge.

Possible results

World model, knowledge graph, causal hypotheses with status and gaps.

Foundation

DINGIR (active research, cooperation with RFI-IRFOS), World / Actor / Causal Model Reconstruction, knowledge graphs.

Maturity and limits

Research-based No published validation result exists. Reconstructions are hypotheses with a status.

Can be combined with

03 Crisis, Failure & Escalation Intelligence04 Forensic & Evidence Intelligence10 Prediction, Scenario & Future Intelligence

Exploratory

Institutional & Socio-Technical Dynamics

How rules, roles, technology and people interact in institutions and change one another.

Institutional & Socio-Technical Dynamics

An institution changes through technology, and it is unclear where dynamics reinforce or block each other.

Use cases

  • Reconstruct the course of escalation in an institutional process from the case files.
  • Study how the use of AI feeds back on processes and roles in an organization.

Possible results

Dynamics map with feedback loops, evidence-based reconstruction, observation recommendations.

Foundation

Space C (Emergence, Adaptation & Complex Dynamics), reconstruction logic, provenance.

Maturity and limits

Exploratory No organizational consulting and no legal assessment.

Can be combined with

03 Crisis, Failure & Escalation Intelligence04 Forensic & Evidence Intelligence09 Emergence & Complex Systems Research

Artificial Intelligence, Systems & TechnologyDevelop, test and embed intelligent systems

Prototype

Multi-Agent Intelligence & AI Engineering

Design agent systems in which roles, handoffs, state and evidence are clearly governed.

Multi-Agent Intelligence & AI Engineering

Several AI units are meant to cooperate, and nobody can trace who is responsible for what.

Use cases

  • Design a specialized agent system for a review or analysis task and develop it into a working system, with roles and traceable handoffs.
  • Examine an existing agent system for questions of responsibility and state.

Possible results

Architecture design, agent system at the agreed maturity level, evaluation report.

Foundation

JARVIS and multi-agent architecture (experimental, cooperation with RFI-IRFOS), runtime, kernel, space E.

Maturity and limits

Prototype The EIL-owned agents are experimental. This describes these instruments and not what can be delivered in an engagement. Production readiness is stated only after evaluation in the specific use.

Can be combined with

09 Emergence & Complex Systems Research14 Adversarial Intelligence, AI Safety & Evaluation15 Research Infrastructure & Intelligence Workbenches

Research-based

Adversarial Intelligence, AI Safety & Evaluation

Test assumptions and claimed system behavior against contradiction, edge cases and failure states.

Adversarial Intelligence, AI Safety & Evaluation

A system was only tested against expected cases and breaks on the first unexpected one.

Use cases

  • Independent review of claims about an AI system against counterevidence.
  • Detect drift and silent loss of previously validated states in an agent system.

Possible results

Review report with contradiction list, test cases and limits of the finding.

Foundation

Adversarial thinking, epistemic integrity, MIRROR (drift and contradiction), JANUS, DAEDALUS.

Maturity and limits

Research-based No certification and no safety proof. A review documents only what was tested.

Can be combined with

13 Multi-Agent Intelligence & AI Engineering04 Forensic & Evidence Intelligence01 Market, Competitive & Strategic Intelligence

Prototype

Research Infrastructure & Intelligence Workbenches

Working environments in which research objects, evidence and decisions stay connected and inspectable.

Research Infrastructure & Intelligence Workbenches

Decisions cannot be traced back to their basis later because context is lost between tools.

Use cases

  • Design the concept of an object-centric working environment for a research team.
  • Develop a visualization of longitudinal data as an interactive research tool.

Possible results

Concept and requirements analysis, prototype or working tool at the agreed maturity level, interactive research visualization.

Foundation

Intelligence Workbench (concept), Observatory (prototype), kernel, space H.

Maturity and limits

Prototype The Workbench is described internally as a concept. What an engagement delivers is set in the proposal by maturity level.

Can be combined with

13 Multi-Agent Intelligence & AI Engineering02 Project & System Intelligence07 Human-AI Interaction & Experience Engineering

Exploratory

AI Transformation, Enablement & Strategic Integration

Enable teams to embed AI systems with an understanding of their limits and effects.

AI Transformation, Enablement & Strategic Integration

An organization introduces AI and wants to know what actually changes as a result.

Use cases

  • Accompanying observation of an AI introduction with measurement points over several months.
  • Workshop format that cleanly distinguishes model capability from system architecture.

Possible results

Observation report, assessment of the introduction, training and handover material.

Foundation

Human-AI co-evolution, LIIE, multi-agent architecture, research protocols.

Maturity and limits

Exploratory No management consulting and no guarantee of efficiency gains.

Can be combined with

07 Human-AI Interaction & Experience Engineering13 Multi-Agent Intelligence & AI Engineering05 Behavioral & Interaction Intelligence

Integrated research architecture

  1. Customer challenge

    Question, goal or development project.

  2. Research configuration

    EIL decides on the basis of goal, data, feasibility and resources what is investigated and what is not.

  3. EIL capability network (16)

    Depending on the engagement, several capabilities take part at once.

  4. Methods, instruments, agents

    Existing procedures, instruments and experimental systems, as far as their maturity allows.

  5. Research, analysis, development

    Execution with documented evidence and disclosed assumptions.

  6. Defined deliverable

    Handover of the results fixed in the proposal.

Example: an engagement on a complex technological development can need strategic intelligence, behavioral intelligence, prediction, world models and adversarial analysis at once. EIL assembles the combination in conversation and in scoping. This describes a way of working and is not an automatic configuration system.

The lab works longitudinally with AI across extended interaction histories, for research, analysis, reconstruction, modeling, system design, experimentation and validation. Isolated prompting plays no part in it. AI is used throughout this process as a research, analysis, modeling and development tool. All concepts, architectures, interpretations and final decisions remain the lab's own responsibility.

What you receive

Every engagement is put together individually. Depending on the question, the outcome can be investigations, decision foundations, models, intelligent systems or new technologies. The tiles show what EIL can deliver within a project. Not all of them belong to every project.

  • Research and intelligence reports

    Research & Intelligence Reports

    Well-founded investigations into complex questions, markets, technologies, people, institutions and system developments.

    Research and intelligence reports

    Research & Intelligence Reports

    Well-founded investigations into complex questions, markets, technologies, people, institutions and system developments.

    You receive

    A structured report with researched information, examined relationships, traceable findings, uncertainties and concrete conclusions.

    Example

    An investigation of new technological developments, their market actors, dependencies, risks and strategic opportunities.

  • Forensic reconstructions and evidence models

    Evidence & Reconstruction Models

    Reconstruction of complex events, contradictory information, decision paths and possible causes.

    Forensic reconstructions and evidence models

    Evidence & Reconstruction Models

    Reconstruction of complex events, contradictory information, decision paths and possible causes.

    You receive

    A traceable chronology, an overview of evidence, a contradiction analysis, hypotheses and, where useful, a visual reconstruction of the relationships.

    Example

    Examine how several decisions and institutional responses led to an escalating situation.

  • Behavioral and interaction studies

    Behavioral & Interaction Studies

    Studies of human and artificial behavior, decision processes, interaction patterns, roles and how they change over time.

    Behavioral and interaction studies

    Behavioral & Interaction Studies

    Studies of human and artificial behavior, decision processes, interaction patterns, roles and how they change over time.

    You receive

    A structured analysis of the observed dynamics, possible explanatory models and documented findings.

    Example

    Examine how collaboration, decisions and roles develop within a team or a human to AI interaction.

  • Future, strategy and forecast models

    Strategic Scenarios & Forecast Assessments

    Examination of possible future developments, influencing factors, dependencies, risks and opportunities.

    Future, strategy and forecast models

    Strategic Scenarios & Forecast Assessments

    Examination of possible future developments, influencing factors, dependencies, risks and opportunities.

    You receive

    Different future scenarios, reasoned assumptions, early indicators and an assessment of possible developments.

    Example

    Which changes could a new technology bring about within a market, and which actors would be affected?

  • Knowledge graphs and world models

    Knowledge Graphs & World Models

    Structured models that connect information, entities, events, relationships, changes over time and dependencies.

    Knowledge graphs and world models

    Knowledge Graphs & World Models

    Structured models that connect information, entities, events, relationships, changes over time and dependencies.

    You receive

    A searchable and, where useful, interactive model that makes relationships visible and information traceable.

    Example

    A world model that links technologies, companies, research developments and their mutual dependencies.

  • Simulations and experimental models

    Simulations & Experimental Models

    Development and study of modeled situations in which different conditions, actors, interactions and possible developments are examined experimentally.

    Simulations and experimental models

    Simulations & Experimental Models

    Development and study of modeled situations in which different conditions, actors, interactions and possible developments are examined experimentally.

    You receive

    A simulation concept or executable model with defined assumptions, scenarios, evaluations and documented limits.

    Example

    Simulate how different decision strategies and interactions could influence the development of a complex system.

  • Intelligent systems and multi-agent technologies

    Intelligent Systems & Multi-Agent Solutions

    Design and development of intelligent systems with specialized AI agents, coordinated decision and execution processes, knowledge integration and individual capabilities.

    Intelligent systems and multi-agent technologies

    Intelligent Systems & Multi-Agent Solutions

    Design and development of intelligent systems with specialized AI agents, coordinated decision and execution processes, knowledge integration and individual capabilities.

    You receive

    Depending on the engagement, a complete system architecture, an intelligence design or a jointly developed, working system, including documentation and agreed quality assurance.

    Responsibilities

    • EIL: Intelligence architecture, multi-agent design, agent logic, behavior models, research foundations and technical evaluation.
    • RFI-IRFOS: Technical implementation, software development and technical integration within the agreed scope of cooperation.

    Example

    A specialized multi-agent system that examines complex information situations, links knowledge, tests hypotheses and produces traceable results.

    The result can be a system ready for productive use. A prototype is the delivery target only when this is agreed explicitly.

  • Independent evaluations and system audits

    Evaluation, Testing & Validation Reports

    Systematic examination of AI behavior, system limits, robustness, errors, drift, interaction quality and the soundness of assumptions.

    Independent evaluations and system audits

    Evaluation, Testing & Validation Reports

    Systematic examination of AI behavior, system limits, robustness, errors, drift, interaction quality and the soundness of assumptions.

    You receive

    A documented test procedure, test results, identified weaknesses, evidence and prioritized options for improvement.

    Example

    Test how a multi-agent system reacts to contradictory instructions, changed contexts and longer interaction sequences.

  • Interactive research tools and visualizations

    Interactive Research Tools & Visualizations

    Development of analysis interfaces, research instruments, interactive models and visualizations of complex information and developments.

    Interactive research tools and visualizations

    Interactive Research Tools & Visualizations

    Development of analysis interfaces, research instruments, interactive models and visualizations of complex information and developments.

    You receive

    Depending on the project, a usable dashboard, an interactive research environment, an analysis instrument or the architecture for one.

    Example

    An intelligence workbench in which events, evidence, agent activity, relationships and changes over time can be examined.

    In joint technical development, EIL takes the research, interaction and intelligence design, and RFI-IRFOS the agreed implementation.

  • Research partnerships and experimental development

    Research Partnerships & Experimental Development

    Joint investigation of new scientific, technical and interdisciplinary questions where solutions or insights arise only in the research process.

    Research partnerships and experimental development

    Research Partnerships & Experimental Development

    Joint investigation of new scientific, technical and interdisciplinary questions where solutions or insights arise only in the research process.

    You receive

    A defined research program with questions, work packages, experiments, interim and final results and, where applicable, newly developed models and technologies.

    Example

    A joint long-term study of human to AI coevolution, or research into new forms of emergent and collective intelligence.

One request, one proposal

  1. Submit your challenge

    You describe the question or goal. You do not have to choose a discipline.

  2. Research scoping

    EIL checks feasibility, requirements and required capabilities.

  3. Individual proposal

    EIL defines scope, results, prerequisites, time frame, price, limits and acceptance criteria. There are no automatic promises of feasibility or price.

  4. Research & delivery

    Execution and handover of the agreed results.

This opens your email app with a prepared message to rfi.irfos@gmail.com. Nothing is sent through our site. You do not have to choose a discipline.