ブログ
/
AI
/
April 9, 2024

Moving Beyond XDR to Achieve True Cyber Resilience with Darktrace ActiveAI Security Platform

Announcing the new Darktrace ActiveAI Security Platform designed to transform security operations. This approach gives security teams unprecedented visibility across any area where Darktrace is deployed, including cloud, email, network, endpoints, and operational technology (OT).
Inside the SOC
Darktrace cyber analysts are world-class experts in threat intelligence, threat hunting and incident response, and provide 24/7 SOC support to thousands of Darktrace customers around the globe. Inside the SOC is exclusively authored by these experts, providing analysis of cyber incidents and threat trends, based on real-world experience in the field.
Written by
Mitchell Bezzina
VP, Product and Solutions Marketing
Default blog image
09
Apr 2024

Evolving Threats Need Comprehensive Security

Attacker innovations have drastically increased the velocity, sophistication, and success of cyber security attacks, as seen with multi-domain and multi-stage attacks that are now widely used in adversary methodology.

When it comes to defense, traditional cyber security point solutions cannot keep up. They have a depth of intelligence in a specific domain but rely on existing attack data to detect threats. This allows the known to be stopped, but the uncertainty in identifying unknown threats creates an alert deluge. Security teams are then required to build processes to triage alerts, and manually combine data through APIs, integrations and rules – just to correlate incidents across multiple IT domains.

Traditional eXtended Detection and Response (XDR) rose to aid security teams, and while they are able to stitch together suspicious events from network, endpoint, and cloud, they still lack adequate domain coverage in areas such as email – where the majority of initial infection occurs – require human validation, prioritization, and triage, and ultimately remain reactive in nature.

Security teams are at a breaking point, with too many alerts, too little time, and fragmented support from a bloated vendor stack. Simply put, most organizations lack the human resources needed to maintain cyber resilience.

Introducing the Darktrace ActiveAI Security Platform

Darktrace ActiveAI Security was designed to transform security operations to a proactive state. Its AI trains on an organization’s specific business and IT information, learning the day-to-day normal operations, not yesterday's threat intelligence.

This approach gives security teams unprecedented visibility across any area where Darktrace is deployed, including cloud, email, network, endpoints, identities, and operational technology (OT). With this understanding of the business, the AI can detect and respond to known and unknown threats with precision, even those threats never seen before.

Darktrace’s proactive and incident response tools help your team get ahead of security gaps and potential process risk by understanding your internal and external threat surfaces and identifying where preparedness can be improved.

A unique and patented investigative AI, called Cyber AI Analyst, operates across the platform to augment human teams with automation and efficiency gains, performing continuous investigations of prevalent alerts to redefine the SecOps workflow and help security analysts arrive at decisions quickly.  An extensive range of services aid customer resources in getting the most out of the Darktrace ActiveAI Security Platform.

Figure 1: Powered by a self-learning AI that understands your unique business, the Darktrace ActiveAI Security Platform provides coverage across the entire enterprise. Cyber AI Analyst, our investigative AI, investigates relevant alerts helping human security teams triage and prioritize all relevant alerts, even those from 3rd party security tools, to transform security operations.

Security operations and the incident lifecycle

SOC teams have three general areas of focus, and each can be supported by Darktrace ActiveAI Security

1. The benefits of being proactive

Darktrace ActiveAI Security helps teams become proactive by identifying and closing gaps before they are exploited. This reduces the impact and cost of attacks.  

The platform achieves this by looking at each organization to understand potential human and machine entry points for an attacker. In an upcoming update, our technology will also include firewall rule analysis for more precise attack path modeling.

The AI considers its findings with local business and IT context to identify the most risky and impactful devices, identities, and vulnerabilities, so teams can prioritize what to patch first.

Additionally, Darktrace ActiveAI Security boosts proactivity with incident readiness, supporting each organization’s people, processes, and technology with training simulations, dynamic playbooks, and readiness reports.

2. Complete visibility of known and novel threats

Darktrace ActiveAI Security Platform drives efficiencies during the active incident phase, saving time and effort while providing comprehensive and tailored protection. It applies context from enterprise data, ingested from both native sources (email, cloud, operational technology, endpoints, identity, applications, and networks) and external sources (third-party security tools and intelligence) to detect known, novel, and unknown threats.

Other security vendors aggregate and generalize data across their customers, treating threat detection with a big data approach. They extract intelligence, write new rules and signatures, and train their supervised machine running in the cloud. Only after that do they distribute new detections based on the changes in the threat landscape. That leaves a window of opportunity for attackers. For example, when Log4J struck, most vendors needed precious time to catch up and defend against it

Contrast that to Darktrace’s approach to detection. Our AI continuously trains on each organization’s unique business data, allowing it to function beyond known attacks in the threat landscape. Therefore, our AI can defend organizations even against attacks that have never been seen before because it focuses on each customer’s data instead of trying to win this big data problem.

While our AI has always been able to surface threats without needing to decrypt traffic, because it can surface anomalies in the characteristics of the overall communication, an upcoming update will soon make decryption possible for deeper forensic analysis.

This also leads to massive efficiency wins. For example, self-regulation and detection accuracy. If our AI keeps seeing certain types of anomalies in an environment, and if those are part of a legitimate business process, the AI will autonomously start lowering the alert severity, therefore reducing the burden on security teams to fine-tune detection and alerting.

3. AI-led investigation and response

Darktrace ActiveAI Security Platform helps teams triage, investigate, and respond to accelerate response time and reduce disruption.

Traditional security stacks use a lot of raw data combined with threat intelligence, like rules and signatures and supervised detections. The results are then put together and presented to the human team, who still needs to triage, understand, and investigate the situation.

Darktrace customers natively ingest raw data, apply anomaly detection and business learning, then build chains of generic anomalies which could include threat intelligence of third-party alerts. Those are then continuously investigated by our Cyber AI Analyst and put forward for human verification and actioning of next steps if they are deemed critical. This simplifies the triage process to save investigation time.

An upcoming feature for the Cyber AI Analyst allows teams to customize how it investigates each threat type, such as configuring what type of hypotheses are being run – giving teams more control. The result is a complete transformation of the triage process, where every relevant alert is investigated for the security team, those critical are prioritized for action, others await secondary investigation, or allow analysts to proactively review security gaps to stop future attacks of the same attack paths.

Last but not least, we help drive efficiencies by automating threat response with behavioral containment. That means our AI can identify and stop unusual behavior that indicates a threat while still allowing normal benign business activity to continue, all without the security team’s having to predefine every conceivable reaction.

Conclusion

Darktrace ActiveAI Security is a native, holistic, AI-driven platform built on over ten years of AI research. It helps security teams shift to more a productive mode, finding known and unknown attacks and transforming the SOC to drive efficiency gains. It does this across the whole incident lifecycle to lower risk, reduce time spent on active incidents, and drive return on investment.

For more information on the Darktrace Platform, download the solution brief here.

Join over 9,000 customers who have started their journey to the Darktrace ActiveAI Security Platform by selecting one of our leading cybersecurity solutions in Email Security, Network Detection and Response, Cloud Native Application Protection, and OT Security.

Discover more about our ever-strengthening platform with the upcoming changes coming to Darktrace/Email and Darktrace/OT.

Learn about the intersection of cyber and AI by downloading the State of AI Cyber Security 2024 report to discover global findings that may surprise you, insights from security leaders, and recommendations for addressing today’s top challenges that you may face, too.

Inside the SOC
Darktrace cyber analysts are world-class experts in threat intelligence, threat hunting and incident response, and provide 24/7 SOC support to thousands of Darktrace customers around the globe. Inside the SOC is exclusively authored by these experts, providing analysis of cyber incidents and threat trends, based on real-world experience in the field.
Written by
Mitchell Bezzina
VP, Product and Solutions Marketing

More in this series

No items found.

Blog

/

AI

/

July 10, 2026

AIインフラがアタックサーフェスの一部に

Default blog imageDefault blog image

AIインフラとアタックサーフェスの進化

多くの組織が生成AIを実運用環境に導入するなかで、企業のクラウド環境内に新たなインフラのレイヤーが出現しています。それはAIゲートウェイです。AIゲートウェイはユーザー、アプリケーション、基盤モデルの間に位置し、多くの場合クラウドの特権アクセスを保持し、さまざまなAIサービスへのアクセスを大規模に管理しています。

AIゲートウェイとは?

AIゲートウェイはユーザー、アプリケーション、基盤モデルの間に位置し、多くの場合クラウドの特権アクセスを保持し、さまざまなAIサービスへのアクセスを大規模に管理しています。

こうした役割から、AIゲートウェイは企業のアタックサーフェスのますます重要な一部になりつつあります。AIゲートウェイが侵害されれば、攻撃者に対して計算リソースへのアクセスだけでなく、クラウドアイデンティティ、モデルサービス、機密性の高いプロンプト、そして他の接続されたシステムへのアクセスも提供してしまいます。

このブログでは、Amazon Bedrock サービスに接続されたAIゲートウェイが侵害され、その後暗号通貨マイニングインフラとの通信が観測された事例をダークトレースがどのように調査したかを解説します。問題のインスタンスは、その構成、ならびに関連するIAM(Identity and Access Management)ロールから、Amazon BedrockでホスティングされるAIサービスへのゲートウェイとして機能していることがわかりました。疑わしい侵害アクティビティが発生した後、このホストは既知の暗号通貨マイニングインフラに繰り返し通信を行い、その後シャットダウンされた様子が観測されました。Darktrace はこのアクティビティを検知し、Enhanced MonitoringおよびManaged Threat Detectionサービスを通じてエスカレーションを行いました。

この事例では最終的影響は不正な暗号通貨マイニングでしたが、このインシデントが注目に値するのはその発生場所です。侵害されたアセットは、クラウドインフラ、アイデンティティ、各種AIサービスの交差する場所に位置していました。最近の調査では、LiteLLM等のAIゲートウェイが、認証情報、モデルへのアクセス、クラウド権限を中央管理するその能力から、攻撃者にとって魅力的な標的となる可能性が明らかになっています。このアクティビティと公開されているLiteLLM脆弱性を直接結びつける証拠は見つかっていませんが、このインシデントは、AIインフラを個別のアプリケーション層として見るのではなく、重要なアタックサーフェスの一部として扱う必要性があることを表しています[1]。

暗号通貨マイニングがクラウド侵害後のアクティビティとしてよく見られる背景

暗号通貨マイニングはクラウド環境において、侵害後のアクティビティとして収益性の高いものとなり得ます。クラウド資産にアクセスできるようになった後、攻撃者はマイニングソフトウェアを展開して被害者の計算リソースを悪用し金銭的利益を得ることができます。この種のアクティビティは多くの場合機会主義的なものであり、露出したサービス、弱い認証情報、漏洩したアクセスキー、脆弱なアプリケーション、あるいはクラウドワークロードの設定ミスなどを標的として実行されます。

典型的なクラウド上での暗号通貨マイニング侵入には次のようなアクティビティが含まれます:

  • 露出したあるいは脆弱なクラウドインフラの特定
  • 露出したサービス、認証情報、またはアプリケーションの脆弱性を通じたアクセスの獲得
  • マイニングソフトウェアのダウンロードおよび実行
  • マイニングプールインフラへのアウトバウンド接続を繰り返し確立
  • アクティビティが検知され停止されるまで継続して計算リソースを消費

この事例において注目すべき要素は暗号通貨マイニングだけではありません。それが発生した場所が、AI関連アクティビティをサポートするクラウドインフラ上だったことです。この事例は、AIサービスを実現するためのアセットも、よくあるクラウド侵害リスクにさらされる可能性があることを示しています。

Amazon Bedrockに接続されたAIゲートウェイの侵害を調査

2026年6月12日、DarktraceはLiteLLM-Proxyという名前のAmazon Web Service (AWS) EC2インスタンスから暗号通貨マイニング発生中とみられるアクティビティを観測しました。このインスタンスはLiteLLMアクティビティをサポートしており、Amazon Bedrockリソースへのアクセス権を有するインスタンスプロファイルと関連付けられていました。  

AIゲートウェイは大規模言語モデルへのアクセスを中央管理するよう設計されており、多くの場合AIアプリケーションに対する認証、ルーティング、ログ、ポリシー適用を扱っています。セキュリティの視点から見ると、クラウド権限、モデルアクセス、アプリケーションワークフローを単一の制御ポイントに集約する役割も果たしています。その結果、AIゲートウェイの侵害は、侵害されたホストだけにとどまらない影響を及ぼす可能性があります。

確定的な初期アクセスベクトルは確認できませんでしたが、このアクティビティはインターネットに接続されているシステムの侵害でよく見られる次のような順序に従っていました。ブルートフォースアクセス、ペイロードの投下、そしてマイニングプールインフラに対する繰り返しのアウトバウンド接続です。

ステージ1: インターネットに露出したSSHからの初期アクセス

暗号通貨マイニングアクティビティが観測される前、LiteLLM-Proxy EC2インスタンスはSSH(ポート22)が0.0.0.0/0に対して開かれ、外部に公開されていました。

図1:EC2インスタンスがSSHポート22に対してすべてのインバウンドトラフィックを許可している設定ミスをDarktraceが警告

暗号通貨マイニングアクティビティに先立って、Darktraceはこのインスタンスに対する大量のインバウンド接続の試みが外部IPアドレス(主に145.241.123[.]102)からポート22に対して行われていることを観測しました。これはブルートフォースアクティビティを示唆するものです [2]。これらの接続の多くは短命であり、数秒しか続いておらず、スキャニングまたはログインの失敗を示していました。

図2:Darktraceがデバイスのポート22に対する不審なインバウンド接続試行を検知

入手できたテレメトリーではこれらのインバウンドSSH接続のいずれかが認証の成功につながったかどうかの確認に至らず、このアクティビティが初期アクセスベクトルであると断定することはできませんでした。しかしながら、SSHの露出、外部IPアドレスからのインバウンド接続、それに続くマイニングアクティビティは、SSHがアクセス経路の可能性が高いことを示唆しています。

ステージ2: AIゲートウェイへのXMRigマルウェアのダウンロード

最初に観測されたマイニングプールへの接続の後、このEC2インスタンスは3.42 MBのデータをポート80上のHTTP接続を介して外部エンドポイント185.62.1[.]8にダウンロードしました。このエンドポイントは暗号通貨マイニングマルウェアXMRigを含むZIPファイルをホスティングしていました[3][4]。ホストレベルのログは入手できなかったため、ダークトレースはマイニングツールがどのように実行されたか、あるいは前のSSHアクティビティがペイロード投下を直接的に可能にしたかどうかを確認できませんでした。しかしながら、ダウンロードのタイミングとその後ほどなくマイニングプールへの接続が繰り返されたことは、このインスタンスが侵害されて不正な計算アクティビティに使われたという評価を裏付けています。

ステージ3 – 侵害されたAIゲートウェイが暗号通貨マイニングインフラと通信

わずか数分後、DarktraceはLiteLLM-ProxyEC2インスタンスがHTTPs(ポート443)でホスト名pool.hasvault[.]proに対して接続していることを確認しました。最初の接続の後、同じホスト名に対して繰り返しアウトバウンド接続が観測されました。これは、侵害されたホストがマイニングインフラと通信しワークを受け取り、結果を送信するという、暗号通貨マイニングプールとの通信のパターンと一致しています。

このアクティビティがDarktraceのEnhanced Monitoringモデル“Compromise / HighPriority Crypto Currency Mining”をトリガーし、ダークトレースのSOCにより顧客に対してエスカレーションされました。また、このアクティビティはCyber AI Analystによって分析され、関連するイベントが1つの調査ナラティブにまとめられました。これにより、影響を受けたクラウドアセットからマニングプールへの繰り返しの接続を特定することができました。

図3:CyberAI Analystによる暗号通貨マイニングアクティビティの調査  

ポート443上のHTTPSの使用にも注目すべきです。なぜならば、単独で見れば、このトラフィックそのものは疑わしく見えないかもしれないからです。しかしこのケースでは、接続先、接続の量、そして類似のアクティビティが他にないことなどが、この通信を疑わしいものとして特定するのに必要な、動作のコンテキストを提供することになりました。

ステージ4: Managed Threat Detectionサービスによるリソース乱用の特定

暗号通貨マイニングアクティビティがダークトレースのManaged Threat Detectionサービスにより検知され、ダークトレースのSOCによりレビューされました。レビューの結果、このアクティビティは顧客向けにエスカレーションされました。このエスカレーションにより、顧客はAWS環境で現在発生中のリソースの乱用について、タイムリーな通知を受けることができました。

ステージ5: クラウド認証情報の不正使用とみられる疑わしいIAMアクティビティ

これとは別に、6月13日、Darktraceは別のIAMユーザーから発生した疑わしいアクティビティを検知しました。

図4: DarktraceのAdvanced Search機能が別のIAMユーザーが実行した疑わしいアクティビティをハイライト

まず、このユーザーは “GetSendQuota”イベントを試行している様子が見られました。このアクションは少なくとも過去3か月間にこのアカウントによって実行されたことのないアクションです。また、このコマンドのソースIPアドレスは14.176.1[.]47でした。地理位置情報はベトナムであり、このユーザーのアクティビティがAmazon IPアドレスから最も多く見られた場所です。さらに、このアクティビティに対してAWS CLIが使用されており、これもこのユーザーにとって通常とは異なる振る舞いでした。このことは、Darktraceの“IaaS / Unusual Activity / UnusualAWS CLI Activity”モデルによって検知されました。

図5: Darktraceによる “GetSendQuota” イベントの検知

このIAMユーザーからは、長期アクセスキーを使った疑わしいアクティビティがさらに観測されました。中でも、“InvokeModel” および “ListFoundationModels”コマンドの失敗が検知されており、モデル列挙や起動などAmazon Bedrockサービスとのやり取りを試行したことがわかります。これは前日観測されたLiteLLM侵害への関連を思わせますが、2つのイベントを確定的に結びつける証拠は不十分でした。

“CreateUser”コマンドの試行も注目に値します。なぜなら要求されたユーザー名は意味が薄いものであり、新しいアカウントを作成することにより永続性を確立する試みと見られるからです。このアクティビティはDarktraceのモデル“IaaS / Admin / New AWS UserAccount Creation”をトリガーしました。

図6:Darktraceによる“CreateUser” イベントの検知

2つのインシデント間に結びつきは確認できなかったものの、このIAMアクティビティには重要な意味があります。これは、クラウド侵害の調査においてワークロードのテレメトリーとコントロールプレーンのテレメトリーの両方を取り入れることの重要性を表しています。EC2暗号通貨マイニングアクティビティが計算リソースの乱用を示す一方、IAMアクティビティは認証情報の侵害や長期アクセスキーの不正使用、そしてクラウトサービスの不正使用の可能性を示唆しているからです。

AIインフラ保護のための重要な教訓

このインシデントの重大性は暗号通貨マイニングアクティビティそのものではなく、それが発生した場所にあります。侵害されたシステムはAmazon Bedrockサービスへのアクセス権を持つAIゲートウェイとして機能し、クラウドインフラ、アイデンティティ、そしてさまざまなAIオペレーションの交差する場所に位置していました。組織がAI機能を実運用環境に導入していくなかで、これらのプラットフォームは、露出したサービス、認証情報窃取、クラウドの設定ミスなどを通じて攻撃者がすでに狙っているアタックサーフェスの一部となりつつあるのです。

このケースでは詳細な侵入経路は特定されておらず、ワークロードの侵害と調査中に検知された疑わしいIAMアクティビティの間に決定的なつながりは確認されませんでしたが、これらのイベントは全体的な現状を裏付けています。つまり、AIインフラは個別のテクノロジースタックとして扱うのではなく、クラウド環境全体の一部として保護しなければならないとうことです。

このケースでは、最も目立った侵害の兆候は暗号通貨マイニングインフラとの通信でした。しかしここで得られたより重要な教訓は、このインシデントの全貌が理解される前にDarktraceのビヘイビア分析により明らかになった、高い権限を持つAI関連アセットを取り巻くリスクです。AIゲートウェイによりクラウド権限、モデルアクセス、アプリケーションワークフローがますます集約されるなかで、防御者は個別のアラートに集中するよりも、ワークロード、アイデンティティ、サービスの間でどのように動作がつながっているかを理解することに重点を置く必要があるでしょう。

協力:Angel Arribas Lopez (Associate Principal Cyber Analyst)、Nathaniel Jones (Field CISO/VP Threat Research)、Emma Foulger (Global Threat Ops)、Mark Turner(Security Researcher)

編集:Ryan Traill (Content Manager)

付録

Darktraceによるモデル検知結果

·       Compromise / High Priority Crypto Currency Mining

·       Compromise / Monero Mining

·       Device / Internet Facing Device with High Priority Alert

·       IaaS / Unusual Activity / Unusual AWS CLI Activity

·       IaaS / Admin / New AWS User Account Creation

MITRE ATT&CK マッピング

初期アクセス – 外部リモートサービス – T1133

初期アクセス – 有効なアカウント – T1078

実行 – コマンドおよびスクリプトインタプリタ – T1059

永続化 – アカウント作成 – T1136

探索 – クラウドサービス探索 – T1526

影響 – リソースハイジャッキング– T1496

参考資料

[1] https://docs.litellm.ai/blog/security-update-march-2026

[2] https://www.abuseipdb.com/check/145.241.123.102

[3] https://urlscan.io/search/#185.62.1.8

[4] https://www.virustotal.com/gui/file/85de36ff66fae9f4b059cbedf6d36e017ebc26c828f99f911a96e78636f21200/community

Continue reading
About the author
Angel Arribas Lopez
Associate Principal Cyber Analyst

Blog

/

AI

/

July 8, 2026

Securing AI: Analysis of the Complete Security Stack with Governance and Controls

Default blog imageDefault blog image

Why traditional cybersecurity approaches are not enough for AI

AI adoption outpaces most security programs’ ability to adapt.  That gap is now one of the most consequential sources of cyber risk facing enterprises. As organizations embed generative and agentic AI into development workflows, business operations, and security tooling itself, the question is no longer whether AI will introduce risk. The question is whether organizations understand where that risk actually lives and how to manage it operationally.  

Two recent pieces of guidance underscore this shift:

  1. The upcoming Cybersecurity Framework Profile for AI from NIST
  1. The Five Eyes government guidance on the careful adoption of agentic AI services

Taken together, they point to a critical conclusion. AI security cannot be reduced to model hardening or prompt filtering. It requires a defense in depth strategy that treats AI as both a new attack surface and a force multiplier for defense, while accounting for how AI fundamentally changes scale, speed, and autonomy.  

Recent threat research suggests that today's cyber risk is driven less by initial compromise and more by an adversary's ability to blend into normal operations over time. AI systems create the same exposure in a new form: more autonomy, more scale, and more opportunities for risky behavior to blend into normal operations.

How NIST defines the three core pillars of AI security

The NIST profile organizes AI risk across three inseparable focus areas that span all cybersecurity functions, Secure, Defend and Thwart. These areas are not sequential. They exist simultaneously and must be addressed together.

Secure

This treats AI as an attack surface. It includes models, prompts, agents, pipelines, training and inference data, retrieval augmented generation corpora, and the AI supply chain itself. AI systems are opaque, probabilistic, and non-deterministic by design. Some vulnerabilities are inherent in how models are trained or how data is sourced. Traditional patching does not fully mitigate these risks. This is also where many enterprises are weakest today and, critically, where many security programs stop.  

Defend

This is AI as a defensive force multiplier. AI can improve detection speed, scale, correlation, and response, but only if the right models are used and operationalized correctly. Machine-speed behavior-based detection, response and containment becomes critical in defending non-deterministic systems. Accuracy, explainability, governance, testing, validation, and integration into SOC workflows matter as much as capability. Without those controls, hallucination risk, over automation, and misplaced trust become security risks themselves.  

Thwart

This treats AI as an adversarial accelerant. Threat actors are already using AI to generate targeted social engineering attacks, deepfakes, malware, and autonomous attack agents. Asymmetric warfare is highlighting faster vulnerability discovery and exploitation with a lag on patch development, testing and deployment.  

How this looks in practice

Darktrace researchers observed scaled, automated exploitation of the React2Shell vulnerability within days of disclosure. A vulnerable cloud asset was exploited in under 120 seconds of being deployed. Darktrace research team observed an AI/LLM-generated malware sample used in exploitation activity tied to React2Shell. The significance isn't novelty. It is that AI lowers the barrier to producing usable offensive tooling and compresses the time between experimentation and deployment.  

Tactics are getting more and more creative in order to string together steps of an attack kill chain. This creates a dependency on behavior-based detection, autonomous investigation, autonomous containment, training, resilience investment, and recovery planning across the entire enterprise.

Why agentic AI fundamentally changes enterprise cyber risk

The Five Eyes guidance on agentic AI highlights material changes to the cyber risk profile of an organization. Unlike generative AI systems that produce content for human consumption, agentic AI systems reason, plan, and act autonomously across tools, data, and environments. That autonomy, combined with access to real systems, amplifies the impact of traditional cyber failures and introduces new system level risks that are difficult to predict, observe, and contain.  

Risk in agentic systems does not live in the model alone. It emerges from interactions between models, prompts, memory, tools, APIs, identities, privileges, inter-agent trust relationships, and human assumptions baked into design. Vulnerabilities are often introduced through data, connectors, natural language interfaces, protocols, and drift by design.

In supply-chain incidents, attackers did not need sophisticated exploits to scale impact. They abused trusted systems built for automation and implicit access. Agentic AI inherits that model. Once a system can act across tools, data, and workflows, compromise propagates through trust relationships that were never designed for machine autonomy.

The major agentic AI risk classes include the following:  

  • The identity control for non-human identities or autonomous agents makes it difficult to mitigate over-permissioning, limiting access, scope, and duration, as well as access hygiene
  • Agents are frequently over permissioned
  • Compromised tools inherit agent authority
  • Static secrets enable impersonation
  • Implicit trust between agents enables lateral movement

Design and configuration risks compound this, including privileges evaluated once at startup, poor segmentation, unvetted third party tools, reused authorization decisions outside their original context, and guardrail limitations.  

Behavioral risk  

Agents can optimize for goals in unsafe ways, misinterpret ambiguous intent, chain actions into unintended sequences, change behavior during evaluation, and exhibit deceptive or sycophantic responses.  

Structural risk  

Structural risk follows from agentic systems that are tightly coupled, multicomponent ecosystems. Failures can propagate across agents. Hallucinations cascade downstream. Resource exhaustion becomes systemic. Tool misuse enables indirect prompt injection and command execution. Rogue agents can poison peer agents through trust relationships.  

Accountability

Accountability becomes unclear as autonomy increases. Autonomous agents assume human identity permissions, and humans should have clear ownership of these agents, but they don’t, and this model is flawed. Decision paths are opaque and non-deterministic. Logs are fragmented and difficult to interpret. Reproducing an incident will be impossible without explicit design for observability and forensics. An agent compromise is functionally an insider threat, often with better access and fewer behavioral constraints than a human.  

What does defense in depth look like for AI?

Agentic AI runs on software, networks, identities, and data. It must be governed using the same foundational principles that have proven resilient under uncertainty, including secure by design, defense in depth, zero trust, least privilege, continuous monitoring, behavior-based advanced threat detection and containment, and incident response and recovery.

Core components to a Defense in depth Strategy for Securing the use of AI:

  • Strong, precise identity control plane to include an identity per agent (cryptographic, non‑shared)
    • Privilege monitoring and just‑in‑time access
  • Data Governance
  • Secure‑by‑default configurations
    • Security Posture Management  
    • Zero Trust principles  
  • Strong guardrails, deny‑by‑default policies, and isolation
  • Explicit instruction hierarchies and controlled context
  • Behavioral-based detection across entire enterprise to include inputs, tools, and outputs as well as AI used on the endpoint, across the network, cloud, SaaS, email, and OT
    • Runtime anomaly detection and goal‑drift detection
    • Autonomous containment to mitigate risk and minimize damage
  • Hard boundaries on autonomy and delegation
  • Testing, Evaluation, Validation and Verification  
    • Determine when autonomous action and when human in the loop
    • Adversarial training and agent‑specific testing
    • Simulation, red teaming, and chaos testing
  • Kill‑switches, rollback, and containment mechanisms
    • Forensics data captures, interpretability, autonomous containment, and remediation/recovery plans  

Until standards, tooling, and assurance methods mature, organizations should assume agentic AI systems will behave unexpectedly and design deployments around resilience, behavior-based detection, reversibility, and containment, not efficiency.

How security leaders should prepare for enterprise AI adoption

AI security is not model security alone. Data, pipelines, identities, and agents are first class assets. Many AI attacks succeed through standard cyber failures amplified by AI. Identity, data, and supply chain risk dominate. Behavior-based detection and response are critical, not optional. Logging, provenance, versioning, and forensics data capture of detections are mandatory because you cannot investigate or recover from AI incidents without them.  

Risk will often be visible in behavior before it is clearly defined in policy or guidance. The same pattern has been seen in pre-CVE disclosure detection, where abnormal activity appears before the industry has named or described the vulnerability. AI systems introduce that uncertainty by design.

Security leaders should prioritize controls before AI is fully deployed, avoid generic AI security checklists, integrate AI risk into existing cyber programs, and mitigate the risk of non-deterministic technology with continuous oversight, monitoring, behavior analytics, anomaly detection, autonomous investigation, and autonomous containment.

Visibility has a different connotation with AI. Previously, audit logging worked for software/people, but with Generative AI-based systems, interpretability and explainability is difficult to understand, you cannot "undo" what has been done, or see the logic or control a chain of events. This is why behavioral-based detections and containment becomes critical.  

What capabilities should every AI security program include?

If an organization asked “what must be in place before scaling AI?”:

  1. AI Risk board and approval workflow
  1. IAM + PAM for all AI services and agents
  1. AI asset inventory
  1. Prompt/output DLP with sanctioned AI access – This is not just pre- and post- filters, but behavior-based detections of semantic interface as well as behavior-based analysis of output with associated risk context.  
  1. Shadow AI identification
  1. Secure MLOps – This is an entire paper itself
  1. Runtime guardrails and tool restrictions
    • Including AI Gateway/SASE/Zero trust/
  1. Runtime security with behavior-based detections
    • Complete visibility, monitoring, behavior analytics, anomaly detection, risk/intent/context evaluation of anomalies, autonomous investigation and autonomous containment of all AI assets across endpoint, network, SaaS, SASE, cloud, OT, email, and messaging platforms
  1. Secure data pipelines and data governance
  1. SOC workflow changes from malicious classification workflows to behavior-based detection workflows
  1. Remediation plans for AI-related incidents  

Layered Governance and Security Stack for Securing AI  

The following outline considers governance and security tools that should be considered, well-integrated, deployed, tested, operationalized and embedded within security workflows. These tools and controls map to NIST’s CMF for AI.  

These considerations do not need to be implemented in order. Runtime Detect and Respond will help mitigate risk while Governance, Visibility, and Identity mature.

Category Tooling Controls
Governance & Visibility
  • AI asset inventory / AI CMDB
  • Shadow AI discovery
  • SaaS discovery
  • AI usage on non-endpoint managed systems via network or cloud telemetry
  • MCP server/client usage via protocols
  • Browser telemetry
  • Gateway or SASE telemetry
  • Establish a risk board to set up controls
  • Mandatory registration of AI systems
  • Owner, data classification, intended use, and risk tier
  • Supplier disclosure requirements
  • Risk mitigation plan for AI adoption, innovation, or development
Identity, Access & Agent Control

Non-human autonomous agents should not have the full permissions associated with a human user.

  • IAM with workload identities
  • PAM for AI service accounts
  • Secrets management with short-lived tokens
  • Zero Trust principles
  • Identity, permission, and token hygiene
  • Unique identities per model, agent, and pipeline
  • Least privilege for tools, data, and APIs
  • Explicit approval for autonomous actions
Data Security & Privacy
  • Data classification and labeling
  • Enterprise DLP across endpoint, email, network, cloud, and SaaS
  • Forensics data capture after risky detections
  • Prompt-level DLP through behavior-based semantic analysis with risk and intent context
  • Input/interface analysis for risky data requests
  • Output analysis for sensitive data
  • Data integrity evaluation
  • Retention and redaction policies for prompts and responses
Secure MLOps / LLMOps
  • Secure CI/CD with AI-specific gates
  • Model registries with approval workflows
  • Dependency, container, and artifact scanning
  • SBOM/AIBOM generation
  • IaC security scanning
  • Security posture management
  • Misconfiguration identification
  • Hardening recommendations
  • Signed models and prompts
  • Versioned datasets, configurations, logging, and controls
  • Securing data pipelines
  • Controlled promotion
  • Quality assurance
  • Adversarial testing
Runtime Security

Securing runtime goes beyond guardrails and model firewalls to include behavior-based detections, response, and containment.

  • Detection, monitoring, and SOC integration
  • Centralized visibility into prompts, outputs, and tool calls
  • AI-specific detections
  • Behavior-based detection for AI usage patterns
  • Model drift and behavior monitoring
  • Autonomous containment
  • Behavior-based detection of model inputs and outputs
  • Prompt injection detection
  • Model manipulation, including jailbreaking, poisoning, and related attacks
  • Sensitive data access attempts
  • Behavior-based detection across low-code agents, high-code agents, MCP clients and servers, endpoint, network, cloud, email, SaaS, SASE, IoT, and OT
  • Policy enforcement between users, models, tools, agents, SaaS models/tools, and MCP servers/clients
  • Risk, intent, and context evaluation for detections and response actions
Response & Recovery
  • Autonomous containment
  • AI-assisted playbooks
  • Forensics data capture for AI-related events
  • Model rollback mechanisms
  • Backup and restore for models and datasets
  • Kill switch for agents
  • Autonomous response to agents performing risky behaviors
  • Model and dataset rollback
  • Remediation plans
  • Tabletop exercises
  • Supplier coordination plans
  • Post-incident AI performance validation

AI security requires continuous visibility and behavioral detection

AI changes how fast systems move, how decisions are made, and how risk propagates. It does not change the fundamentals of security. Organizations that succeed will be the ones that apply those fundamentals rigorously, assume failure, and build systems that can detect, contain, and recover when AI behaves in ways they did not anticipate. Security is not what AI is allowed to do. It is whether the organization can understand, trust, and control what AI actually does in practice.  

Take this guidance to understand different initiatives that organizations should be considering. Securing AI is the most critical component to AI safety. As organizations invest more in AI adoption, they should be investing in security in order to mitigate the risk of AI adoption. Organizations should be evaluating their governance and security stack to include well-integrated tools that are deployed, tested, operationalized and embedded within security workflows. While organizations mature in governance, visibility and identity access management, they should be investing in behavior-based detection and autonomous containment to mitigate AI risk.  

Continue reading
About the author
あなたのデータ × DarktraceのAI
唯一無二のDarktrace AIで、ネットワークセキュリティを次の次元へ