REPOSITORY_ID: AXL-RSCH-DB 02_RESEARCH
RSCH_02 [ THE_RESEARCH_AGENDA ]

Built like a university hospital.

We're built like a university hospital: the commercial lab earns at market rates and funds the nonprofit side, which runs the public-good research.

FIG. 02 // ONE MISSION, TWO BODIES
For-profit · Aux Labs LLC The commercial lab

Clients pay market rates for research, creative and policy work.

On its own: a working practice that pays its own way.
Nonprofit · being stood up The Switchboard™

Runs the trust-repair cascade as grant-funded public-interest work.

On its own: grant-funded, run on the published protocol.
Like a teaching hospital: the clinic pays for the research, and the research makes the clinic better. Money crosses the firewall; influence over findings does not.

[ FIELD_NOTE ]The field is behavioral infrastructure. The method is applied mechanism design: the economics discipline of Hurwicz, Maskin and Myerson. They are not the same thing, and we do not collapse them.

Behavioral Infrastructure Applied Mechanism Design Causal Inference Computational Social Science
FLAGSHIP_PROGRAM

The Pretend Era™

The lab's flagship research program is a falsifiable framework diagnosing institutional trust collapse as a pluralistic-ignorance equilibrium: a condition in which most people privately reject the official story but comply with it, believing everyone else accepts it. The framework specifies measurable signatures and the conditions under which it would be wrong. Everything else the lab builds is downstream of this diagnosis.

Read AXL-WP-04, the public edition ->

WORK_WITH_US

Like how we think? Bring us a problem.

Every paper here started as someone's live problem. If yours looks like one of them, let's talk.

[ BOOK_A_MEETING ]

Or write to imran@auxlabs.co

PRIOR_PUBLICATIONS
Peer-Reviewed

Jin, M.X., Rajan, S., Gary Bicas, C.E., Hao, M., Dong, L., Mufson, B., & Hafiz, I. (2021). Novel Validated Index for the Measurement of Disinformation Susceptibility at the County Level. Cureus, 13(5). -> Known as the Jin-Hafiz Index. PubMed · PMID 34211810

THE_WORKING_PAPER_INDEX // AXL-WP-01 … AXL-WP-11

Swipe or use the arrows

A studio microphone with handwriting in several scripts running through it.

Borrowed Voices: Stigma, Preference Falsification, and the Human Data Problem in AI Alignment

Request an advance copy ->
INSTRUMENTS // THE RESEARCH, BUILT INTO TOOLS
An iron plumb bob hanging true on its cord; behind it, Mission Control’s consoles, out of focus. INSTRUMENT_01 // AI OVERSIGHT Is anyone still checking the AI? An anonymous instrument that measures whether human oversight of an AI deployment still works. Its headline number, the candor gap, is the Pretend Era from AXL-WP-04, measured inside one organization. SEE_PERPENDICULUM ->
FIELD_PLANS // WORK THAT RUNS OUTSIDE THE INDEX
A woven straw beehive filled with honeycomb. THE COMMONS Social media stopped being social. A credible plan to move people off a broken social media model and into communities that are actually social, rooted in trust and consent. The first one is already running. READ_THE_NECTAR ->
ROUTES // WHO THIS PAGE IS FOR

Academics.

A testbed with real deployment for questions that usually die at the end of a paper. Method named, data open. Cite us.

Grant Funders.

Predictions stated in advance, and a structure built so funding cannot bend the finding.

Journalists.

The data is open and the firewall is written down. Conflicts disclosed before you ask.