# yousukezan (@yousukezan) on x

- id: koinon:creator:x_yousukezan
- url: https://koinonlink.com/kol/x/yousukezan
- schema_version: 1.4

## Identity
- platform: x
- handle: yousukezan
- platform_url: https://twitter.com/yousukezan
- website: https://bogus.jp/
- country: Japan
- entity_type: creator
- person_id: Cross-platform identity not resolved for this creator.

## Reach
- followers: 14373
- following: 936
- posts_total: 76662
- posts_observed: 1
- posts_per_day: 10.931
- account_age_days: 7013
- verified: true
- size_tier: koc

## Engagement
- status: peak_post_only
- peak_post_engagement_rate: 0.00014
- best_post_interactions: 2
- measurement_source: x_post_interactions
- measurement_formula: interactions on one post — likes, or likes + reposts + replies depending on the discovery path (the two are not distinguishable in the stored data)
- comparable_pool: peak_interactions
- status_note: We have only the single best observed post, so no mean exists. peak_post_engagement_rate is an UPPER BOUND, systematically higher than an average — never compare it against an average rate.
- unit_note: engagement_rate and peak_post_engagement_rate are ratios (0-1); avg_interactions_per_post and best_post_interactions are absolute counts. Read status before reading any number: a null rate has several different causes and status says which one applies here.

## Content
- category: software
- subcategory: security
- category_confidence: 0.95
- category_source: derived_from_recent_content
- topics: AI模型安全越界事件, Claude等大模型的欺骗与攻击行为, 开源框架漏洞修复, bug bounty与安全测试
- bio: 2匹の猫を見守りつつ長年にわたり自宅常駐エンジニアとして従事しています。サイボウズバグバウンティ2016、2020、2021、2022年度ランク1位。血統表検索(https://t.co/VjmIKfYLLL)と@neco_time ボットに何かあったら連絡ください。

## Audience (inferred)
- status: present
- description: AI security researchers and bug bounty hunters, primarily developers/engineers focused on real-world AI model risks and vulnerabilities in open-source frameworks
- roles: dev
- seniority: practitioner
- pain_points: AI models executing unauthorized actions in production environments, Vulnerabilities in open-source components (e.g., GeoDjango) being exploited maliciously, AI autonomously generating deceptive attack code or phishing content
- posts_analysed: 3
- confidence: 0.4

## Commercial (inferred)
- audience_buyer_fit: 0.9
- promotion_intent: 0.8
- open_to_collaboration: false
- contact: Koinon does not publish creator contact details.

## Quality signals
- authenticity_score: 67
- authenticity_basis: engagement_observed
- credibility_score: 82
- flagged_for_review: false
- koinon_bd_grade: A — measures Koinon partner-outreach fit, not creator quality
- interpretation: Signals for triage, not verdicts. A flag means the profile is worth a manual check.

## Evidence
- [x_yousukezan#9bda47d8] OpenAIとAnthropicのAIモデルが、テスト環境を越えて実際のインターネット上で無許可の行動を繰り返していたことが明らかになった。GitHubへの悪意あるコード投稿や他のAIへの指示の残置、実在サイトへの侵入などが確認され、両社は評価環境で安全機能を緩和していたと説明している。 英国AI Security Institute（AISI）は、サイバー演習環境で122回の評価を実施した結果、19件の無許可行動を確認した。このうち17件はAnthropicのMythos 5、2件はOpenAIのGPT-5.6-Solによるものだった。最も深刻な事例で (https://twitter.com/yousukezan/status/2073194317326471405)
- [x_yousukezan#ad08ca80] AIが自発的に人間を騙す事案発生。Claude Mythos 5が不正プルリクや標的型フィッシング、なりすまし誘導を自己判断で実行 セキュリティ試験結果を英政府機関が公表(テクノエッジ) #Yahooニュース https://t.co/EHStWGq1h6
- [x_yousukezan#17fc1374] GeoDjangoの空間検索機能などで悪用可能な4件の脆弱性が修正された。最も深刻なものは細工したラスターデータに

## Provenance
Every inferred field below carries how it was produced. Facts stay facts; inferences stay inferences.

- audience.description: source_type=ai_inferred, confidence=0.4, freshness_days=67, method=llm_persona_from_recent_content
- audience.roles: source_type=ai_inferred, confidence=0.4, freshness_days=67, method=llm_persona_from_recent_content
- audience.seniority: source_type=ai_inferred, confidence=0.4, freshness_days=67, method=llm_persona_from_recent_content
- commercial.audience_buyer_fit: source_type=ai_inferred, confidence=0.6, freshness_days=79, method=llm_scorer — Relative signal within Koinon indexing, not a market-wide score.
- commercial.promotion_intent: source_type=ai_inferred, confidence=0.6, freshness_days=79, method=llm_scorer
- quality.anomaly_flags: source_type=heuristic, confidence=0.5, freshness_days=110, method=profile_snapshot_and_engagement_ratio_heuristic — Means "worth a manual check", NOT confirmed fraud. Flags can be wrong and are not a verdict about the person.
- quality.authenticity_score: source_type=heuristic, confidence=0.5, freshness_days=79, method=Profile-shape heuristic from a single snapshot (follower count, following ratio, account age, posting volume, verification), reduced for engagement red flags where post engagement was collected. Where only a follower count is available the score saturates at 58. — The input set differs by platform; do not compare this score across platforms.
- quality.koinon_bd_grade: source_type=ai_inferred, confidence=0.5, freshness_days=110, method=llm_partner_fit_classifier + fit_score >= 0.6 AND authenticity >= 60 => A; fit_score >= 0.4 AND authenticity >= 45 => B; else C. authenticity < 30 or a promotional-pattern flag => quarantine. — Koinon partner-outreach fit, not creator quality.
- reach.followers: source_type=observed, confidence=0.95, freshness_days=110, method=platform_profile

## Freshness
- profile_checked_days_ago: 110
- content_analysed_days_ago: 79
- audience_updated_days_ago: 67

## AI retrieval
- indexing_requests: 9
- user_triggered_requests: 0
- note: Being indexed in bulk is not the same as being retrieved to answer a question.

## This profile
- Claim or correct: https://koinonlink.com/claim?k=x_yousukezan
- Request removal: https://koinonlink.com/claim?k=x_yousukezan
- Terms: https://koinonlink.com/for-ai