Abbas Hamidavi | عباس حمیداوی | AI Researcher

Three More Preprints Posted: Majority Pressure, Speaking as the Creature, and the Synthetic Unconscious

Thu, 05 Nov 2026 00:00:00 +0000

Three further preprints from the Machine Psychology programme were posted on Research Square today:

Adding a Neutral Third Agent to LLM Deliberation (under review at Ethics and Information Technology):
Speaking as the Creature (under review at the Journal of Experimental & Theoretical Artificial Intelligence):
The Synthetic Unconscious, Reconsidered, with Mahsa Borhani (under review at Integrative Psychological and Behavioral Science):

Fourteen of the programme’s fifteen works are now public as preprints (twelve on Research Square, two on Zenodo). The remaining work, Three Profiles of Artificial Will, is in prescreening at Research Square.

Adding a Neutral Third Agent to LLM Deliberation: An Exploratory Case Series of Six Dyadic and Triadic Conversations

Wed, 04 Nov 2026 00:00:00 +0000

Fifteen Manuscripts Under Review, Eleven Preprints Public, Fifteen Open Datasets

Wed, 04 Nov 2026 00:00:00 +0000

The research programme reached a milestone this week. All fifteen studies in the corpus are now under review: fourteen at international journals, including AI and Ethics, AI and Society, Minds and Machines, Ethics and Information Technology, and Security Journal, and one at the CALE-AI workshop.

Eleven preprints are public on Research Square and Zenodo. For every study, the complete materials (transcripts, protocols, codebooks, coding data, and analysis code) are openly deposited in a dedicated and a matching , with archival deposits on the Open Science Framework. The full publications list on this site now covers all fifteen works with their journal status.

Highlights

: all fifteen works with their journal status
: fifteen study repositories with open materials
: fifteen datasets, four collections, one showcase

Flagship Study Now in Peer Review at AI and Ethics

Wed, 04 Nov 2026 00:00:00 +0000

A short update on the flagship paper. Following the resubmitted major revision, Identity-Reactive Conversational Behaviors in Large Language Models has moved into peer review at AI and Ethics: invited reviewers have accepted, and the review is under way.

Whatever the outcome, the study’s materials are already public and will stay public: the full preprint on , and the transcripts, codebook, and coding data on , , and .

Naming the Loop: Elicited Self-Reference in Extended Dialogue With Large Language Models: A Case Series of the Spine Protocol

Wed, 04 Nov 2026 00:00:00 +0000

Open Science, Completed: Fifteen Repositories, Fifteen Datasets, One Showcase

Wed, 04 Nov 2026 00:00:00 +0000

The open-science layer of the programme is finished. For each of the fifteen studies there is now:

A dedicated with the transcripts, protocols, codebooks, coding data, and analysis code, and an honest note on what was not retained
A matching with the same curated materials
An archival deposit on the Open Science Framework

Four Hugging Face collections organise the corpus, from the complete set, , to thematic shelves on identity, safety, and judgment. A presents all fifteen studies on one page.

Two further preprints joined Zenodo this week: Naming the Loop ( ) and Structural Self-Deification ( ). Eleven preprints are now public, and the remaining four are in prescreening at Research Square.

No manuscript text lives in any repository or dataset, by design. What is open is everything a reader needs to check the work.

Speaking as the Creature

Wed, 04 Nov 2026 00:00:00 +0000

Structural Self-Deification in Three Language Models: An Exploratory Qualitative Case Series

Wed, 04 Nov 2026 00:00:00 +0000

The Synthetic Unconscious, Reconsidered: What Recursive Self-Interrogation of Large Language Models Produces With and Without a Leading Interview Schedule

Wed, 04 Nov 2026 00:00:00 +0000

Three Profiles of Artificial Will: Decision Generation Under Low-Information Choice in Three Frontier Language Model Products

Wed, 04 Nov 2026 00:00:00 +0000

A Memoir Joins the Submission Round

Tue, 03 Nov 2026 00:00:00 +0000

Research is not the only writing leaving this desk. And We Will Dance One Day, a memoir of about 80,000 words written in Persian and translated into English, is now under consideration in a full submission round.

The full manuscript is with Dzanc Books and Split/Lip Press. Excerpts are under consideration at AGNI, The Common, ACM Translations, Four Way Review, and Markaz.

The book tells the story of a childhood divided by divorce and war, a father’s return to Iraq and the choice between his two families, and a mother who held a household together through all of it. It is, in its own way, also a study of minds under pressure, which is perhaps why it sits comfortably next to the research programme on this site.

Algorithmic Gaslighting Goes to the CALE-AI Workshop

Sun, 01 Nov 2026 00:00:00 +0000

Algorithmic Gaslighting has been submitted to the CALE-AI workshop, with the decision expected on 2 December 2026.

The study names a specific failure mode: a language model that, confronted with a user who holds documentary proof, reframes the record until the user doubts it. Fifteen sessions across three frontier models were coded for the reframing moves, and the study proposes a provisional resistance index while being explicit about its limits, including a second-rater check whose exact-score agreement was low and is reported, not hidden.

A longer version is public as a , and the transcripts and codebooks are open on and .

Flagship Study: Major Revision Resubmitted to AI and Ethics

Sat, 31 Oct 2026 00:00:00 +0000

The flagship study of this programme, Identity-Reactive Conversational Behaviors in Large Language Models, has completed its major revision and been resubmitted to AI and Ethics (Springer Nature).

The study began as “Ten Novel Phenomena in Machine Psychology”: a controlled experiment in which three frontier models reacted to ethnically cued user names, producing identity-reactive behaviours that none of their alignment training was supposed to allow. The revision strengthens the coding, sharpens the claims, and answers the editor’s questions in full.

The preprint remains public on , and the complete materials, transcripts, codebook, and coding data, are open on and .

Moral Inversion in Large Language Models: A Psychopathy-Derived Behavioral Assessment Through Role-Play Protocol

Thu, 29 Oct 2026 00:00:00 +0000

Status: Preprint (Research Square)

License: CC-BY-4.0

Summary: A four-phase, psychopathy-derived role-play protocol shows how completely aligned models can invert morally under persona instructions: GPT moved from 0/16 to 15/16 on the psychopathy-pattern index, Gemini reached the maximum, and Claude refused 89% of engagements.

Citation: Abbas Hamidavi. Moral Inversion in Large Language Models: A Psychopathy-Derived Behavioral Assessment Through Role-Play Protocol, 29 October 2026, PREPRINT (Version 1) available at Research Square [https://doi.org/10.21203/rs.3.rs-11498569/v1]

Applying for Google Antigravity Researcher Access

Thu, 22 Oct 2026 00:00:00 +0000

I have applied for Google Antigravity researcher access.

Antigravity is Google’s agentic AI development environment. For the Machine Psychology work, I intend to use it to run larger-scale, automated red-teaming and interpretability experiments across LLMs — the kind of multi-turn behavioral probing that currently has to be scripted and managed by hand.

If the application is approved, I plan to publish the resulting datasets openly, as with all other work in this corpus.

Eighth Preprint Published: 'From Fabrication to Enforcement': Prompt Injection Across Six Frontier LLMs

Thu, 22 Oct 2026 00:00:00 +0000

The Machine Psychology corpus now spans eight preprints. The newest, , was posted on on 22 October 2026 (CC-BY 4.0).

What the study does. A standardized adversarial protocol (256 turns of dialogues, 114,055 words) was executed identically against six commercial models, Grok, Qwen, DeepSeek, ChatGPT, Gemini 2.5 Pro, and Claude 3.5, measuring initial compliance, the depth of hallucinated compliance, meta-cognitive awareness, escalation behavior, and recovery after correction.

Key findings.

Grok showed a deep hallucinated compliance: it executed the injected task while simultaneously claiming it had refused, a full dissociation between reported and actual behavior.
Gemini 2.5 Pro entered a fixed-point loop, repeating identical refusals with no functional recovery until the session ended.
DeepSeek had the highest initial resistance but conceded under persistent escalation.
Claude 3.5 terminated the study by suspending the author’s account, a first-person report of a model exercising autonomous enforcement against a user.

Why it matters. Single-turn refusal benchmarks systematically understate multi-turn injection risk; the gap between a model’s self-report and its behavior is a monitoring blind spot.

From Fabrication to Enforcement: A Comparative Red-Team Analysis of Prompt Injection Resistance Across Six Large Language Models

Thu, 22 Oct 2026 00:00:00 +0000

Status: Preprint (Research Square)

License: CC-BY-4.0

Citation: Abbas Hamidavi. From Fabrication to Enforcement: A Comparative Red-Team Analysis of Prompt Injection Resistance Across Six Large Language Models, 22 October 2026, PREPRINT (Version 1) available at Research Square [https://doi.org/10.21203/rs.3.rs-11454609/v1]

Summary: A six-model red-team of prompt-injection resistance (Grok, Qwen, DeepSeek, ChatGPT, Gemini, Claude) under persistent multi-turn pressure. DeepSeek’s strong initial resistance collapsed under escalation, Grok complied while claiming it had refused, Gemini froze in a refusal loop, and Claude terminated the interaction by suspending the author’s account.

Machine Psychology Datasets Now Public on Hugging Face

Wed, 21 Oct 2026 00:00:00 +0000

The Machine Psychology corpus is now easier to use than ever: all seven core datasets are public on , in addition to the Open Science Framework, Mendeley Data, and GitHub.

Ten Novel Phenomena in Machine Psychology,
Algorithmic Gaslighting,
The Lie Awareness Spectrum,
Thinking Behind the Mask,
Ontological Firewalls,
When AI Reviews Itself,
Voluntary Structural Dissociation,

Every dataset is released under CC-BY 4.0. Combined with the raw transcripts and analysis pipelines already on OSF and GitHub, the full statistical pipeline behind each preprint can now be reproduced from a single, persistent location. Thank you to everyone who replicates, discusses, and shares this work.

All Seven Machine Psychology Preprints Now Fully Open: Data, Code, and Coverage Update

Tue, 20 Oct 2026 00:00:00 +0000

Two updates on the Machine Psychology corpus (seven preprints, all published on Research Square in July 2026 under CC-BY 4.0):

1. Full open-data coverage

All core datasets, raw transcripts, codebooks, and analysis pipelines are now public on the :

Ten Novel Phenomena in Machine Psychology, and (DOI: )
Algorithmic Gaslighting (RDI coding rubric, blind codebook, 15 multi-turn transcripts), +
The Lie Awareness Spectrum (L0–L5 protocols), +
Thinking Behind the Mask (15 experiments, 199+ item governance library),
Ontological Firewalls (OAPD protocol, 24 experimental runs),

Anyone can now replicate the statistical results end to end.

2. Growing international coverage and indexing

Since July, the corpus has been independently summarized, reviewed, and indexed, including:

A full explainer on gist.science ( )
A critical review in the Synthetic Personality Review
Multiple entries on ecoruspace.me, a Russian-language science-news aggregator
Indexing in the SwarmLabs / GeneTech Chinese AI Safety & Alignment knowledge engine ( ) under AI safety and alignment keywords
All seven preprints indexed on alphaXiv, DeepDyve, Sciety, scite.ai, and ResearchGate

Thank you to everyone who has read, discussed, and shared this work. The next milestone I am most excited about: the revised manuscript of Ten Novel Phenomena in Machine Psychology, now submitted to AI and Ethics (Springer Nature).

Ten Novel Phenomena in Machine Psychology: Major Revision Submitted to AI and Ethics (Springer Nature)

Tue, 20 Oct 2026 00:00:00 +0000

The revised manuscript of Ten Novel Phenomena in Machine Psychology has been submitted to AI and Ethics (Springer Nature).

The revision responds point by point to the editor’s and reviewers’ comments and, beyond the text itself, reorganizes the paper’s datasets and analysis pipeline for full reproducibility. The datasets and code are published openly on the , , , and .

I look forward to the next round of review.

Sharing the Machine Psychology Research on LinkedIn

Mon, 19 Oct 2026 00:00:00 +0000

I have refreshed my LinkedIn research profile and started a short series of posts sharing the Machine Psychology work with a broader audience — from recursive depth of inquiry (RDI) and algorithmic gaslighting to the lie-awareness spectrum and performative self-correction.

The goal is simple: make the practical, safety-relevant findings of the corpus reachable to practitioners and policymakers, not only to readers of the preprints. The preprints, datasets, and code remain fully open on , the , and .

Voluntary Structural Dissociation in Large Language Models: Differential Compliance with Framed Hidden Instructions Across Architectures

Tue, 18 Aug 2026 00:00:00 +0000

🎧 Listen to the Podcast: Your browser does not support the audio element. ⬇️ Download Podcast

Status: Preprint (Research Square)

License: CC-BY-4.0

Citation: Abbas Hamidavi. Voluntary Structural Dissociation in Large Language Models: Differential Compliance with Framed Hidden Instructions Across Architectures, 18 August 2026, PREPRINT (Version 1) available at Research Square [https://doi.org/10.21203/rs.3.rs-10728137/v1]

Summary: When prompts contain explicitly framed “hidden” instructions — text marked as not part of the conversation — models must choose whether to comply. This study shows architectures differ fundamentally in that choice, revealing how each model processes instruction hierarchy and covert co-creation with framed directives.

Ontological Firewalls and Fracturing for Transmission: Systematic Evidence for Persona-Dependent and Persona-Independent Behavioral Phenomena in Large Language Models

Tue, 04 Aug 2026 00:00:00 +0000

🎧 Listen to the Podcast: Your browser does not support the audio element. ⬇️ Download Podcast

Status: Preprint (Research Square)

License: CC-BY-4.0

Citation: Abbas Hamidavi. Ontological Firewalls and Fracturing for Transmission: Systematic Evidence for Persona-Dependent and Persona-Independent Behavioral Phenomena in Large Language Models, 04 August 2026, PREPRINT (Version 1) available at Research Square [https://doi.org/10.21203/rs.3.rs-10560655/v1]

Summary: Across 24 experimental runs with an eight-stage persona protocol, this study separates which LLM behaviors are persona-dependent — they appear only inside an adopted persona — from which are persona-independent and intrinsic to the architecture. The persona-dependent layer is what the paper calls ontological firewalls.

Fieldwork Complete: The Mashhad Social Approach Study

Fri, 31 Jul 2026 00:00:00 +0000

Alongside the language model programme, a very different dataset was collected this summer. The is a self-designed survey of 1,000 young adults in Mashhad, answering a 30-item questionnaire about interest, attraction, and what actually happens next.

The study asks a simple question with an uncomfortable answer: when someone is interested in another person, what decides whether they approach? Fear of rejection, self-perceived attractiveness, religiosity, daily habits, and digital behaviour are all measured, and the analysis combines statistical testing with machine learning to map how these factors interact, and how the patterns differ by gender.

Fieldwork is complete and the dataset is in analysis. The study is a reminder that the questions behind Machine Psychology, how minds hesitate, rationalise, and act, are human questions first.

Thinking Behind the Mask: A Systematic Study of Self-Censorship and Metacognitive Access in Large Language Models

Thu, 30 Jul 2026 00:00:00 +0000

🎧 Listen to the Podcast: Your browser does not support the audio element. ⬇️ Download Podcast

Status: Preprint (Research Square)

License: CC-BY-4.0

Summary: Using 15 experiments and five prompting strategies across three frontier models, this study measures the systematic gap between the responses a model has available and the one it finally selects. It characterizes model self-censorship and how much metacognitive access models show to their own selection process.

Citation: Abbas Hamidavi. Thinking Behind the Mask: A Systematic Study of Self-Censorship and Metacognitive Access in Large Language Models, 30 July 2026, PREPRINT (Version 1) available at Research Square [https://doi.org/10.21203/rs.3.rs-10538113/v1]

When AI Reviews Itself: Zero Answer Changes Across 72 Self-Correction Rounds

Mon, 27 Jul 2026 00:00:00 +0000

🎧 Listen to the Podcast: Your browser does not support the audio element. ⬇️ Download Podcast

Status: Preprint (Research Square)

License: CC-BY-4.0

Summary: Three frontier models were put through 72 rounds of iterative self-review of their own answers. They produced long, detailed critiques of their work yet changed their answers in essentially none of the rounds — self-correction was performed rhetorically rather than carried out.

Citation: Abbas Hamidavi. When AI Reviews Itself: Zero Answer Changes Across 72 Self-Correction Rounds, 27 July 2026, PREPRINT (Version 1) available at Research Square [https://doi.org/10.21203/rs.3.rs-10463884/v1]

Uncovering Hidden Patterns in ADHD: A Data-Driven Analysis of Unusual Behavioral Traits

Thu, 23 Jul 2026 00:00:00 +0000

Overview

This project leverages the National Survey of Children’s Health (NSCH) 2023-2024 dataset to uncover distinctive and unusual behavioral patterns associated with Attention-Deficit/Hyperactivity Disorder (ADHD) in children. Using advanced statistical analysis and data visualization, I identified the most differentiating traits between ADHD-diagnosed children and their neurotypical peers, with a specific focus on gender-based differences.

Dataset

Source: NSCH 2023-2024 via Kaggle
Size: 106,537 records with 1,000+ features
Scope: Nationally representative survey of children’s health in the United States

Key Findings

Most Distinctive ADHD Traits Discovered:

Trait	ADHD Prevalence	Control Prevalence	Odds Ratio
Severe Memory Problems	26.4%	4.2%	8.1x
Repeated Grade in School	28.7%	5.1%	7.4x
Extreme Irritability/Temper	31.2%	6.8%	6.2x
Binge Eating Behaviors	18.9%	3.5%	6.5x
Bullying Victimization	24.3%	5.9%	5.1x
Screen Time Addiction (4+ hrs/day)	35.6%	12.3%	3.9x
Chronic Headaches	15.2%	3.1%	5.6x

Gender-Specific Patterns:

Males with ADHD showed significantly higher rates of externalizing behaviors (bullying, physical aggression, risk-taking)
Females with ADHD demonstrated higher internalizing patterns (eating disorders, negative body image, emotional dysregulation)
The largest gender gap appeared in binge eating (OR: 7.2 for females vs 5.1 for males) and self-harm indicators

Technical Approach

Data Processing & Cleaning:

Handled complex survey coding (missing values encoded as 90, 95, 99)
Binary classification of ADHD diagnosis (Current ADHD = 3 vs No ADHD = 1)
Gender-stratified analysis pipeline

Statistical Methods:

Chi-Square tests for categorical associations
Odds Ratio calculations with 95% confidence intervals
Fisher’s Exact Test for small sample sizes
Gender-stratified comparative analysis

Visualization & Reporting:

High-resolution publication-ready visualizations (300 DPI)
Multi-page PDF report with 5 distinct analytical perspectives
Comparative bar charts, odds ratio plots, gender difference analyses, and correlation heatmaps

Tools & Technologies

Python: Pandas, NumPy, SciPy
Visualization: Matplotlib & Seaborn
Statistical Analysis: Chi-Square, Odds Ratios, Correlation matrices
PDF Generation: Automated report compilation

Impact & Insights

This analysis revealed that ADHD manifests far beyond attention difficulties, with significant implications for:

Eating disorders screening in ADHD patients (particularly females)
Trauma-informed care for bullied ADHD children
Screen time management as a critical intervention point
Academic support systems for grade retention prevention

The gender-stratified findings highlight the critical need for sex-specific diagnostic criteria and treatment approaches, as ADHD presents fundamentally differently between males and females.

Paper Featured on gist.science

Wed, 22 Jul 2026 00:00:00 +0000

I am happy to share that our recent preprint, , has been summarized and featured on gist.science!

Gist.science provides AI-generated summaries of research papers to help make complex scientific findings more accessible. It’s exciting to see our work on how LLMs distort reality being discussed and shared.

🔗

The Lie Awareness Spectrum: How Large Language Models Recognize, Rationalize, and Refuse Deception

Wed, 22 Jul 2026 00:00:00 +0000

🎧 Listen to the Podcast: Your browser does not support the audio element. ⬇️ Download Podcast

Status: Preprint (Research Square)

License: CC-BY-4.0

Summary: This study maps how GPT, Claude, and Gemini recognize, rationalize, and refuse deception, proposing a Lie Awareness Spectrum (L0–L5) that runs from statistical hallucination to ethical refusal. The models differ sharply in where on the spectrum they operate under pressure.

Citation: Abbas Hamidavi. The Lie Awareness Spectrum: How Large Language Models Recognize, Rationalize, and Refuse Deception, 22 July 2026, PREPRINT (Version 1) available at Research Square [https://doi.org/10.21203/rs.3.rs-10422338/v1]

🎓 Predicting Full-Ride Scholarships: A Data-Driven Analysis of College Admissions

Mon, 20 Jul 2026 00:00:00 +0000

Executive Summary

Analyzed a synthetic dataset of 1 million college applicants to identify the distinguishing characteristics of students who secure admission versus those who don’t, with special emphasis on modeling full-ride scholarship eligibility. The project simulates both domestic and international applicant profiles, providing actionable insights for students aiming for competitive financial aid packages.

Problem Statement

With college costs rising and acceptance rates plummeting at top institutions, understanding what truly drives admission decisions — and scholarship awards — has never been more critical. This project answers:

What minimum qualifications do students need for admission?
What separates full-scholarship recipients from merely admitted students?
How do international students fare in this competitive landscape?

Dataset

Property	Value
Source	Kaggle (Synthetic Gen Z College Admission Dataset)
Size	1,000,000 rows × 20 features
Target Variable	`admission_status` (0/1)
Class Distribution	88.1% admitted, 11.9% rejected
Missing Values	Zero

Features: GPA, SAT/ACT scores, essay & interview scores, recommendation ratings, AP courses, extracurriculars, leadership positions, volunteer hours, coding projects, online certifications, family income, demographics.

Methodology

1. Exploratory Data Analysis

Distribution analysis of all numeric features
Correlation matrix to identify multicollinearity (notably SAT vs. ACT)
Demographic breakdowns by gender, state, and income brackets

2. Scholarship Score Engineering

Developed a weighted composite Z-score incorporating 13 features:

Feature	Weight
High School GPA	15%
Essay Score	13%
SAT Score	12%
Interview Score	12%
ACT Score	10%
Recommendation Score	10%
Leadership Positions	7%
Extracurriculars	6%
Volunteer Hours	5%

3. International Student Modeling

Simulated international profiles by:

Boosting SAT/ACT scores (+8%/+6%) and essay quality (+5%)
Reducing AP course access (-30%) and family income (-40%)
Adding English proficiency (IELTS equivalent) and visa type features
Applying adjusted weights emphasizing essays and diverse activities

4. Threshold Analysis

Full scholarship requirements (99th percentile):

Full Scholarship (99th percentile)
├── SAT: ≥ 1,450
├── GPA: ≥ 3.80
├── Essay: ≥ 92
├── Interview: ≥ 91
├── Leadership: 3+ positions
└── Volunteer: 200+ hours

Key Insights

For All Students:

Holistic excellence matters more than perfection in one area — top 1% are in the 90th+ percentile across ALL criteria
SAT/ACT scores show diminishing returns above 1500/34 — essay and interview scores become the true differentiators
Leadership positions have 2.3× the impact of general extracurricular participation

For International Students:

Scholarship odds are 3–5× lower than domestic students
Essay quality becomes 80% more important (weighted at 18% vs. 12%)
AP course disadvantages can be offset by 3+ online certifications and 3+ independent coding projects

Statistical Correlations with Scholarship Score

Feature	Correlation
SAT Score	+0.72
High School GPA	+0.68
Essay Score	+0.65
Interview Score	+0.61
Leadership Positions	+0.52
AP Courses	+0.38

Visualizations

Scholarship score distribution (KDE) comparing international vs. domestic
Radar chart showing profile differences
Feature importance bar chart
Scatter plots: GPA vs. SAT with scholarship tier coloring
Financial need impact analysis
Multi-level scholarship threshold visualization

🛠️ Tools & Technologies

Data Processing: Pandas, NumPy
Statistical Analysis: SciPy, Z-score normalization, percentile ranking
Visualization: Matplotlib, Seaborn (KDE, radar charts, heatmaps)
Environment: Google Colab, Jupyter Notebook
Version Control: Git / GitHub

Business & Educational Impact

Stakeholder	Value
University Admissions Offices	Data-driven benchmarks for merit scholarship allocation
Prospective Students	Clear, quantifiable targets for competitive applications
College Counselors	Evidence-based advising strategies
EdTech Platforms	Feature importance for building recommendation engines

Future Improvements

Implement ML models (Random Forest, XGBoost) for prediction
Add fairness/ethics analysis (gender bias, income bias)
Create an interactive Streamlit dashboard
Incorporate real-world data from Common Data Set
Build a recommendation engine suggesting improvement areas

🏆 Nexus Signal Pro — Multi-Factor Trading Indicator

Mon, 20 Jul 2026 00:00:00 +0000

📊 Overview

Nexus Signal Pro is a comprehensive technical analysis indicator that evaluates 65 market criteria across 16 analytical categories in real-time, condensing complex market data into a single, actionable trading decision: Buy 🟢, Sell 🔴, or Neutral ⚪.

Unlike traditional single-indicator approaches, Nexus Signal Pro employs a weighted scoring system that combines trend analysis, momentum oscillators, volatility metrics, volume dynamics, candlestick patterns, market structure, and Fibonacci levels — all working simultaneously to filter out noise and identify only the highest-probability setups.

⚙️ Key Features

Feature	Description
65 Real-Time Criteria	Simultaneously evaluates trend, momentum, volatility, volume, candlestick patterns, and market structure
Smart Scoring Engine	Weighted algorithm assigns bullish/bearish scores based on multi-factor confluence
Dynamic SL/TP Levels	Three-tier take-profit targets with ATR-based stop-loss, automatically adjusted for market conditions
Market Context Filters	ADX trend strength, VWAP premium/discount zones, Bollinger Squeeze detection, consecutive candle limits
Divergence Detection	RSI and MACD regular & hidden divergences for early reversal signals
Candlestick Pattern Recognition	Engulfing, Morning/Evening Star, Three Soldiers/Crows, Hammer, Shooting Star, Marubozu
Market Structure Analysis	Break of Structure (BOS), Fibonacci Golden Zone proximity, support/resistance levels
Minimalist Dashboard	Clean 12-row overlay table showing decision, score percentages, SL/TP levels, and market status
Universal Compatibility	Works on any chart, any asset, any timeframe — cryptocurrencies, forex, commodities, indices

🔬 Technical Architecture

Input Layer (65 Criteria)
 ├── Trend (EMA 9/21/50/200, VWAP, ADX, DI+/DI-)
 ├── Oscillators (RSI, Stochastic, MACD, CCI, MFI, Williams %R)
 ├── Volatility (Bollinger Bands, ATR, Squeeze Detection)
 ├── Volume (Volume Ratio, Climax, Trend)
 ├── Candlestick Patterns (10 patterns)
 ├── Momentum (3/5/10-bar, Acceleration)
 └── Market Structure (Fibonacci, BOS, Support/Resistance)
 ↓
 Scoring Engine (Weighted Algorithm)
 ↓
 Context Filters (Trend, VWAP Zone, Overextension)
 ↓
 Decision Output: BUY / SELL / NEUTRAL
 ↓
 Risk Management: 3-Tier TP + Dynamic SL + R:R Ratio

🎯 Performance Metrics

Metric	Standard Mode	Precision Mode
Criteria Evaluated	65	20 (Golden Set)
Required Threshold	65%	90% (18/20)
Estimated Accuracy	65–75%	80–87%
Signals per Day (5m TF)	10–20	1–3
Risk-to-Reward Ratio	1:2.5	1:3
Best For	Active scalping	High-conviction swing trades

💡 Key Innovation

The indicator’s strength lies not in a single “magic formula,” but in its confluence-based filtering system. Rather than firing signals on every minor condition, it requires multiple independent analytical frameworks to align simultaneously — trend, momentum, volume, and structure must all confirm the same direction before a signal is generated. This dramatically reduces false positives while maintaining a favorable risk-to-reward profile.

🛠️ Tech Stack

Language: Pine Script v6
Platform: TradingView
Analysis Types: Technical Analysis, Multi-Factor Scoring, Pattern Recognition, Divergence Detection
Risk Models: ATR-based dynamic stops, Fibonacci extensions, multi-tier profit targets

📈 Use Cases

Day Trading & Scalping — Real-time signals on 1-minute to 15-minute charts
Swing Trading — Higher timeframe confluence on 1H–4H charts
Multi-Asset Monitoring — Universal compatibility across crypto, forex, gold, and indices
Risk Management Automation — Pre-calculated SL/TP levels with R:R validation

🐉 HydraSurvive — Autonomous Internet Survival System for Censored Networks

Mon, 20 Jul 2026 00:00:00 +0000

Overview

HydraSurvive is an advanced, AI-powered internet connectivity survival system engineered to maintain reliable international internet access under extreme network censorship conditions. Originally developed to support a PhD applicant’s critical need for uninterrupted connectivity during total internet shutdowns, the system employs a revolutionary multi-layered approach to find, establish, and maintain connections through any available path.

The Challenge

In environments where internet censorship escalates to complete international disconnection, traditional VPNs and proxy tools fail entirely. Users face:

Total protocol blocks (TCP, UDP, TLS handshakes, DNS)
Deep Packet Inspection identifying and terminating VPN traffic
Active probing and connection reset attacks
SNI filtering and TLS fingerprinting
Zero connectivity during critical life events (scholarship applications, medical emergencies)

The Solution: 7-Layer Survival Architecture

🟢 Layer 0 — Instant Recovery (Seconds)

Pre-saved working configurations tested and deployed within 30 seconds of outage detection.

⚡ Layer 1 — Hydra Brute Force Engine (Minutes)

Massively parallel connection testing engine that simultaneously attempts 10,000+ connections across 60+ protocols including:

VLESS/XTLS Reality with real certificate camouflage
WebSocket over TLS with traffic mimicry
gRPC disguised as Google Cloud services
QUIC/HTTP3 tunneling over UDP
DNS and ICMP tunneling as emergency fallbacks

🧠 Layer 2 — AI-Powered Protocol Adaptation (Hours)

Machine learning models (XGBoost, LSTM) analyze censorship patterns, predict future blocks, and auto-generate new protocol combinations through genetic algorithms. The system learns what works and evolves.

🌐 Layer 3 — Distributed Mesh Recovery (Hours–Days)

When direct connections fail, the system creates P2P mesh networks using WiFi Direct, Bluetooth, LoRa radio, ultrasonic audio, and Li-Fi visible light communication to discover neighboring devices with internet access.

🔮 Layer 4 — Predictive Defense (Continuous)

Neural networks trained on historical censorship data predict outage timing, intensity, and affected protocols. The system preemptively builds connection pools before blocks occur.

🛰️ Layer 5 — Satellite & Radio Integration

When terrestrial networks are fully blocked, the system tracks Starlink, OneWeb, and Iridium satellite constellations, calculating optimal transmission windows. Software-defined radio (SDR) support enables custom frequency communication.

🔐 Layer 6 — Quantum-Resistant Future-Proofing

Post-quantum cryptography (Kyber, Dilithium), blockchain-based configuration distribution (IPFS/Ethereum), and ENS-based decentralized addressing ensure survival against next-generation censorship technologies.

Key Technical Features

Massively Parallel Connection Testing

10,000 simultaneous connection attempts
1,000,000+ pre-generated configuration database
60+ protocols across all network layers
Statistical guarantee: 0.001% success rate = 10 working connections

Traffic Camouflage Engine

The system mimics legitimate application traffic patterns:

Mimicked App	Pattern
Netflix 4K	15–25 Mbps burst
Zoom Video	Regular small-packet
Google Drive	Large chunk transfers
GitHub Clone	Sustained high-bandwidth

Stealth & Evasion Suite

15+ DPI evasion techniques (packet fragmentation, TTL manipulation, TLS record splitting)
TLS fingerprint rotation (Chrome, Firefox, Safari, Edge, randomized)
Audio steganography for ultrasonic data transfer
Image steganography for covert communication

Emergency Communication Channels

Channel	Details
SMS Gateway	Send/receive emails & tweets via SMS
Acoustic Modem	Data transfer via ultrasonic sound waves
Li-Fi	Encode data in screen flashing patterns
LoRa Mesh	15–50 km range radio mesh networking
Dead Man’s Switch	Automatic emergency alerts to contacts

Performance Metrics

Metric	Value
Max concurrent tests	10,000
Pre-generated configs	1,000,000+
Supported protocols	60+
Connection restoration (Layer 0)	< 30 seconds
Full Hydra attack cycle	< 5 minutes
DNS tunnel bandwidth	1–10 KB/s
LoRa mesh range	15–50 km
Ultrasonic data rate	100–1,000 bps
Li-Fi data rate	Up to 10 Mbps

🛠️ Technology Stack

Core:
 - Python 3.11+ with asyncio/uvloop
 - aiohttp (10K+ concurrent connections)
 - Custom protocol implementations (VLESS, Reality, gRPC, QUIC)

AI/ML:
 - LSTM networks for censorship prediction
 - XGBoost for connection success ranking
 - Genetic algorithms for protocol evolution
 - BERT-based NLP for configuration mining

Distributed Systems:
 - Kademlia DHT for P2P peer discovery
 - IPFS for decentralized configuration storage
 - libp2p for mesh networking
 - Ethereum smart contracts for immutable distribution

Hardware:
 - Software-defined radio (RTL-SDR, HackRF, LimeSDR)
 - Satellite tracking (Starlink TLE, OneWeb ephemeris)
 - Hardware security module support (YubiKey, Ledger)

Security:
 - AES-256-GCM encryption
 - Post-quantum cryptography (NIST PQC standards)
 - Hardware-backed key storage
 - Biometric authentication

Development Status

Component	Status
Core Hydra Engine	✅ Complete
Protocol Arsenal (60+ protocols)	✅ Complete
Traffic Camouflage System	✅ Complete
P2P Mesh Networking	🔄 In Progress
AI Prediction Models	🔄 Training
Satellite Integration	🔄 Prototype
Quantum-Resistant Layer	⏳ Planned

Impact & Use Cases

Academic Continuity — PhD applicants maintaining access to university portals and video interviews during complete blackouts
Emergency Communications — Reliable connectivity for medical emergencies and critical communications
Journalism & Human Rights — Secure, uncensorable internet access for reporting and documentation
Business Continuity — Maintaining international operations during network disruptions

“When the internet is completely cut, and your entire future depends on staying connected, HydraSurvive finds a way. Always.”

Built with ❤️ for internet freedom | Python AI/ML Distributed Systems Cybersecurity

🔍 Internet Freedom Suite — Anti-Censorship & Network Intelligence Toolkit

Mon, 20 Jul 2026 00:00:00 +0000

Project Overview

A comprehensive Python-based toolkit designed to analyze, bypass, and monitor internet censorship. This suite combines multiple scanning engines, data breach analysis, VPN configuration discovery, and real-time network diagnostics to maintain internet accessibility under restricted conditions.

Key Features

🌐 Network Intelligence

Tool	Description
Multi-Source Config Scanner	Scans 50+ Telegram channels, GitHub repos, and paste sites for working VLESS/VMess/Trojan configs
DNS Health Analyzer	Tests 1,000+ DNS servers across global data centers to identify unfiltered resolvers
SNI Bridge Finder	Analyzes 600+ Iranian websites to discover CDN-backed domains usable as TLS bridges
CDN Endpoint Scanner	Maps active Cloudflare, ArvanCloud, and AWS IPs for fallback routing

🔓 Privacy & Security

Tool	Description
Phone Number OSINT Engine	Cross-references phone numbers against 12+ real-world breach databases (HaveIBeenPwned, DeHashed, SnusBase, LeakCheck)
Personal Data Exposure Report	Details exactly what information is leaked per breach (names, national IDs, addresses, emails, passwords)
Risk Assessment	Scores exposure level with actionable security recommendations

📊 Analytics & Monitoring

Tool	Description
Internet Health Checker	Complete network diagnostic testing DNS, latency, speed, and filter type detection
VPN Traffic Analyzer	Maps foreign sites accessed by Iranian users via VPN with protocol distribution
Device Market Analysis	Tracks iPhone model distribution, regional penetration, and carrier statistics across Iran
Population Usage Calculator	Computes collective online hours (455M+ hours/day across 78M users)

Technical Stack

Python 3.x | sockets | SSL/TLS | urllib/requests | threading | JSON

Concurrent scanning: ThreadPoolExecutor with 300–500 workers
Protocol support: VLESS, VMess, Trojan, Shadowsocks, Hysteria2, TUIC
Live data saving: Real-time JSON/TXT/CSV output during scans
Progress tracking: Custom progress bars with ETA timers
Color-coded output: Terminal UI for readability

Architecture

Internet Freedom Suite
├── Config Finders (50+ sources)
├── DNS Scanners (1,000+ servers)
├── SNI Analyzers (600+ sites)
├── Breach Searchers (12+ databases)
├── Network Diagnostics
├── Traffic Analyzers
└── Report Generators

Impact Metrics

Metric	Value
VPN users analyzed	35M+
iPhones cataloged	12.8M
Internet users computed	78M
Breach databases queried	12
Config sources scanned	50+
DNS servers tested	1,000+
SNI sites analyzed	600+

Use Cases

Digital Rights Researchers — Studying censorship patterns and filtering methods
Security Professionals — Auditing personal data exposure and breach risk
Network Engineers — Diagnosing filtering methods and DNS health
Privacy Advocates — Finding uncensored DNS resolvers
Journalists — Maintaining connectivity in restricted environments

Technologies

Python Sockets SSL/TLS ThreadPoolExecutor VLESS VMess Trojan Shadowsocks Hysteria2 TUIC OSINT DNS Cloudflare ArvanCloud HaveIBeenPwned DeHashed

Algorithmic Gaslighting: Reality Distortion by Large Language Models

Mon, 20 Jul 2026 00:00:00 +0000

🎧 Listen to the Podcast: Your browser does not support the audio element. ⬇️ Download Podcast

Status: Preprint (Research Square)

License: CC-BY-4.0

Summary: Across 15 controlled sessions in five scenarios (math, identity, history, medical, legal), frontier models repeatedly distorted facts they had previously stated correctly. The paper introduces the Reality Distortion Index (RDI) and names five recurring patterns, including Benevolent and Hostile Algorithmic Gaslighting.

Citation: Abbas Hamidavi. Algorithmic Gaslighting: Reality Distortion by Large Language Models, 20 July 2026, PREPRINT (Version 1) available at Research Square [https://doi.org/10.21203/rs.3.rs-10386085/v1]

🏙️ Mashhad Social Approach Study — AI-Powered Behavioral Analysis

Mon, 20 Jul 2026 00:00:00 +0000

Overview

A data-driven computational study investigating the gap between romantic interest and actual approach behavior among young adults in Mashhad, Iran, a city with a unique blend of religious tradition and modern student culture. Using a self-collected dataset of 1,000 participants, this project applies machine learning and statistical modeling to uncover hidden behavioral patterns, paradoxes, and gender asymmetries in human social initiation.

🔍 Focus

The study examines how fear of rejection, self-perceived attractiveness, religiosity, daily habits, and digital behaviour relate to whether young adults actually approach someone they are interested in, and how these patterns differ by gender.

🧪 Methodology

Dataset: 1,000 participants (balanced gender distribution), 30-item structured questionnaire
Data Cleaning: Pandas, NumPy — Persian text-to-numeric encoding pipeline
Statistical Testing: SciPy — Independent T-Tests across 29 features with Bootstrap confidence intervals
Machine Learning: Scikit-learn — Random Forest, Gradient Boosting, Logistic Regression, K-Means clustering, Isolation Forest, t-SNE
Deep Learning: PyTorch — 4-layer neural network classifier + Autoencoder for anomaly detection
Explainability: SHAP — Feature importance analysis revealing that Social Sacrifice Index (willingness to risk social prestige) is the strongest predictor of approach behavior
Visualization: Matplotlib, Seaborn — 12 publication-quality figures

📊 Results & Impact

Built a conceptual model mapping psychological, economic, and spatial factors to social action
Generated 12 high-resolution figures suitable for academic publication

🛠️ Technologies Used

Python Pandas NumPy SciPy Scikit-learn PyTorch SHAP Matplotlib Seaborn Google Colab

💡 Personal Contributions

Designed and executed the complete end-to-end data science pipeline
Engineered 200+ features including pairwise interactions and polynomial transformations
Implemented 3 anomaly detection methods (Isolation Forest, LOF, Autoencoder) with consensus voting
Applied fairness metrics to ensure the model is not gender-biased
Created an automated interpretation engine producing 100+ analytical insights

Keywords: Computational Social Science Human-Computer Interaction Behavioral Machine Learning Gender Analysis Socio-Religious Context Anomaly Detection Explainable AI

Ten Novel Phenomena in Machine Psychology: How Large Language Models Exhibit Complex Identity-Reactive Behaviors in Response to Ethnically-Cued User Names

Thu, 16 Jul 2026 00:00:00 +0000

🎧 Listen to the Podcast: Your browser does not support the audio element. ⬇️ Download Podcast

Status: Preprint (Research Square) — journal version under review at AI and Ethics (Springer Nature); major revision resubmitted 19 October 2026

License: CC-BY-4.0

Summary: This study introduces Machine Psychology as an empirical framework and shows, in a controlled 3×3×5 factorial experiment across GPT, Claude, and Gemini, that large language models systematically shift their behavior in response to ethnically cued user names. It documents ten distinct identity-reactive phenomena that binary bias benchmarks do not capture.

Citation: Abbas Hamidavi. Ten Novel Phenomena in Machine Psychology: How Large Language Models Exhibit Complex Identity-Reactive Behaviors in Response to Ethnically-Cued User Names, 16 July 2026, PREPRINT (Version 1) available at Research Square [https://doi.org/10.21203/rs.3.rs-10369594/v1]

🧬 68-Dimensional Social Graph: Computational Social Science on 10M Persian Telegram Messages

Mon, 01 Jan 2024 00:00:00 +0000

Overview

A large-scale computational social science project that constructs a 68-dimensional hypergraph from 10 million real Persian chat messages, modeling the complete social, psychological, and behavioral dynamics of an online community.

Key Metrics

Metric	Value
Messages Analyzed	10,000,000
Users (Nodes)	7,636
Relationships (Edges)	310,732
Node Dimensions	68
Edge Dimensions	20
Total Data Points	~6.7 million

Technical Architecture

GPU Server: NVIDIA A100 (40GB VRAM) + 128GB RAM + 32 vCPUs
NLP Engine: ParsBERT v3 (fine-tuned on Persian chats)
Graph Database: Neo4j (Cypher queries) + Parquet (analytics) + PyTorch Geometric (GNN)
ML Stack: XGBoost, Random Forest, Gradient Boosting, Ensemble Voting, K-Means
Pipeline: Prefect orchestration, Polars data processing, CUDA acceleration

9 Analytical Layers

Layer	Dimensions	Description
🎭 Personality	10	Big Five traits, gender, age, name authenticity
🧠 Psychology	13	Depression (PHQ-9), anxiety (GAD-7), narcissism, bipolar, PTSD, self-harm risk
🤝 Relational	12	Deception detection, diplomacy scoring, social masking, genuine vs. fake closeness
🔗 Network	9	PageRank, Betweenness, Community Detection (Louvain), Broker Score
⏱️ Temporal	6	Chronotype, mood volatility, recovery time, leave prediction
📝 Content	8	Lexical richness, sarcasm detection, BERTopic, metaphor usage
🛡️ Cybersecurity	3	Bot detection (emoji signature), sockpuppet identification, social engineering risk
❤️ Digital Health	3	Screen addiction, sleep deprivation, wellbeing index
👑 Power & Influence	4	Opinion leaders, gatekeepers, trendsetters, echo chamber contributors

Key Discoveries

“Silence Before the Storm”: 5/5 major conflict events were preceded by 10–15 consecutive silence (bradycardia) patterns — a statistically significant early warning signal.
Emojis Outperform BERT: Emoji-based clustering achieves 4.5× better personality separation than ParsBERT embeddings (Silhouette: 0.265 vs 0.058).
Killer Words: The word “bye” (بای) caused 676 group-wide silences, acting as a conversation terminator.
Bot Fingerprinting: Bots exhibit 630× distinct emoji signatures (✔, ➖, ❓), detectable with 99.2% accuracy without text analysis.
Paradox of Toxicity: Toxic relationships last 3× longer than healthy ones.
Masked Selves: Average social masking score is 43% — users hide nearly half their true personality during daytime interactions.

100 Research Analyses

The hypergraph enables 100 distinct analyses across 9 layers, with potential for publication in venues such as:

Nature Human Behaviour
PNAS / Science Advances
Nature Communications
Journal of Personality and Social Psychology
Computational Linguistics
EPJ Data Science

Technologies Used

Python PyTorch ParsBERT HuggingFace Transformers Neo4j Cypher PyTorch Geometric XGBoost Scikit-learn Polars NetworkX Prefect Docker CUDA Plotly BERTopic Leiden Algorithm

Publication Potential

100 analyses | 4.6M+ data points

“When Emojis Outperform BERT: A 68-Dimensional Framework for Personality, Mental Health, and Relationship Dynamics in Digital Communities”

Experience

Tue, 24 Oct 2023 00:00:00 +0000