Labs & Research

Public prototypes
and experiments.

AULAS LATAM’s lab: we explore emerging technologies with clear hypotheses, publish what we learn and only turn into products what gets validated.

5
Research lines
4
Stages per experiment
Research lines

Five fronts,
one method.

Advanced AI

Agents, retrieval and evaluation

Agents with scoped tools and permissions, RAG over owned sources and systematic evaluation of quality, cost and risk before any deployment.

AgentsRAGEvalsHuman review
01
Line
Open
Collaboration
Model security

AI systems that withstand attacks

Prompt injection, data leakage, tool abuse and red teaming, using the OWASP Top 10 for LLM Applications as the reference framework.

Prompt injectionRed teamingGuardrailsOWASP LLM
02
Line
Open
Collaboration
Post-quantum cryptography

Getting systems ready for the quantum era

Cryptographic inventory, crypto-agility and testing with NIST standards: ML-KEM (FIPS 203), ML-DSA (FIPS 204) and SLH-DSA (FIPS 205).

ML-KEMML-DSASLH-DSACrypto-agility
03
Line
Open
Collaboration
Blockchain

Verifiable trust without intermediaries

Verifiable credentials for certifications, process traceability and tamper-evident records, always checking whether a conventional database is enough.

Verifiable credentialsTraceabilityW3C VC
04
Line
Open
Collaboration
Emerging tech

What’s next, tested with judgment

On-device AI, WebGPU, open data and new interfaces. We test early to decide with evidence, not hype.

On-device AIWebGPUOpen data
05
Line
Open
Collaboration
Experiment board

What we’re
testing now.

Every experiment starts with a question. If the answer is “no”, that gets published too.

Line
LAB-01Exploration

Intelligent document analysis

Classify and extract document data with per-field traceability and a human review queue.

Question
What per-field precision is reachable without human review?
LAB-02In development

Semantic AI glossary

Meaning-based search, in Spanish and English, over the Lexia glossary (Glosario IA by Tu Aula Latam).

Question
Does semantic search beat literal search at finding terms?
LAB-03Exploration

Prompt-injection test bench

Reproducible attack cases to evaluate agents that read email, documents and web pages.

Question
Which combination of defenses lowers the attack success rate most?
LAB-04Exploration

Cryptographic inventory

Detect quantum-vulnerable algorithms in repositories and configurations.

Question
How much of a typical app depends on RSA and ECC?
LAB-05Exploration

Verifiable certificates

Issue “Certificación IA para el Trabajo” certificates as credentials third parties can verify.

Question
Can an employer verify a certificate in seconds without calling anyone?
LAB-06Prototype

Public procurement radar

Connector to SECOP II open data to analyze technology procurement in Colombia.

Question
Which signals anticipate an organization’s technology need?
LAB-07Exploration

Adaptive microlearning tutor

Adjust pace and difficulty of short lessons from each learner’s answers on TuAula.

Question
Does one-week retention improve versus a fixed path?
LAB-08Exploration

In-browser AI

Small models running on-device with WebGPU for features that keep data off servers.

Question
Which tasks are viable on a mid-range phone?
Lab method

Hypothesis, prototype,
evaluation, publication.

An experiment ends when it answers its question, not when time runs out. It may become a product on aulas.lat, a capability on aulas.si, a course on tuaula.lat or a report on what we learned.

  1. Hypothesis

    A measurable question and the criterion that would answer it.

  2. Prototype

    The minimum needed to get evidence, with test data and no real personal data.

  3. Evaluation

    Metrics defined upfront, with risks and limitations documented.

  4. Publication

    Hypothesis, method, results and limitations, even when the result is negative.

Got a hypothesis
worth testing?

Researchers, graduate students and specialists can propose experiments or join one in progress.