ZEISMO
Research & lab
The methods lab

A better starting point.

Free, practical tools to help plan an experiment, document what happened and preserve an inspectable record.

AI Impact ObservatoryExplore twelve study programmes on business, financial markets and public policy: two published computational pilots and ten proposed studies. Read the programme and methods →
01 / EXPERIMENT PROTOCOL

Define the question first.

Download an editable JSON protocol with fields for hypotheses, datasets, baselines, exclusions, metrics, repeated runs and ethical considerations.

Question → Hypothesis → Method
Datasets & permissions
Baselines & evaluation metrics
Exclusions & repeated runs
Analysis plan & limitations

Save a dated protocol before collecting data and document later amendments.

02 / EXPERIMENT RUN LOG

Record every attempt.

A CSV template for consistent run records, including model versions, seed, configuration, outcome, failures and resource costs.

run_id, timestamp_utc, model_version
dataset_version, configuration, seed
metric_name, metric_value, outcome
failure_reason, runtime_seconds, cost

Keep failed and excluded runs, with reasons. Do not add private customer information or secrets to shared logs.

03 / REPRODUCIBILITY CHECKLIST

Make the method inspectable.

Track your experimental preparation with a checklist saved on this device.

0 of 8 preparation steps checked.

04 / FILE INTEGRITY

Fingerprint your record.

Calculate a file’s SHA-256 checksum locally, then compare it with the fingerprint stored in our publication archive.

Exact contents → SHA-256 fingerprint
Same contents → Same fingerprint
Changed contents → Different fingerprint
Open the integrity tool

Compare the exact file contents against the archived version using its SHA-256 fingerprint.

From idea to evidence

Keep the reasoning visible.

01

Plan

Write a testable question and decide what would count as supporting or contradicting evidence.

02

Run

Preserve configurations, unsuccessful attempts and measurable outcomes.

03

Challenge

Look for data leakage, selection bias, weak baselines and alternative explanations.

04

Share

Present methods and limitations so someone else can understand and test your claims.

Develop your research

Have a method worth examining?

Prepare a structured manuscript, document your methods and submit your work for editorial review.

Open your workspace
The ZEISMO AI programme

From executable code
to coordinated decisions.

Read the architecture report

Led by Engineer Mohanad Osama Zahran, the programme investigates how an AI system can combine specialised skills, contextual evidence and memory to carry out increasingly capable roles.

01 / INTERFACE

JARVIS

The proposed English-speaking interface turns a request into a defined task and communicates the system’s response, uncertainty and next step.

02 / COORDINATION

System Brain

The research architecture coordinates role-specific agents, evidence and permissions. It separates a proposed decision from authority to execute it.

03 / SPECIALISATION

Skills and cells

Each proposed software cell combines a role contract, curated skills and scoped tools. Cells represent software responsibilities, not biological neurons.

04 / ADAPTATION

Memory with evidence

Proposed memories retain their source, time and review state. Candidate lessons are tested before promotion into a role’s working procedures.

Research horizon / 100 specialists

What would a hundred-specialist system need to prove?

Our ambition is to coordinate complementary specialist capabilities and exceed human performance on defined tasks. The research must measure decision quality, completion time, cost, disagreement and unsafe actions against a specified human comparison under equivalent conditions.

“100 times smarter than a human” is a research ambition, not a measured result. The published JARVIS report examines a prototype and controlled simulations; the follow-on decision study defines 48 synthetic scenarios. Neither establishes a general human-intelligence multiplier.

Explore the decision dataset

From routine work to AI employees

We investigate the phased replacement of suitable routine tasks with role-specific AI employees. Each expansion needs evidence about task quality, approval boundaries, exception handling and recovery.

Context before action

In this programme, “feel” describes appraisal of operational context: changing conditions, incomplete evidence, risk and consequences. It does not establish consciousness or subjective experience.

Read JARVIS & System BrainMeet the research leadExplore AI employee research
ZEISMO research practices

Evidence you can inspect.

Follow each marker to the method, record or policy behind it.

Versioned recordsIdentifiers and integrity checks Documented methodsProtocols and reproducibility tools Inspectable evidenceMethods, aggregates and fingerprints Accountable publishingAuthorship and correction policies

ZEISMO practice markers. Verification scope is recorded with each publication.