One Article Review
| Source |  ProofPoint |
|---|---|
| Identifiant | 8648557 |
| Date de publication | 2025-02-10 01:58:04 (vue: 2025-02-15 01:11:26) |
| Titre | AI in Cybersecurity: the Good, the Bad and the Ugly |
| Texte | Artificial intelligence (AI) is revolutionizing cybersecurity. It offers unparalleled capabilities for detecting, predicting and neutralizing threats in real-time. But at the same time, threat actors are using it to create sophisticated attacks. At Proofpoint, we take an evidence-based, practical approach to AI. Our Nexus® AI Framework combines advanced machine learning with behavioral analytics for explainable decision-making. As a result, you get actionable insights and protection across multiple threat vectors. This blog explores the many sides of AI-the good, the bad and the ugly. It also details how the six AI cores in Proofpoint Nexus power our targeted solutions, which stop today\'s most complex security challenges. Proofpoint Nexus AI: defending from every angle Proofpoint Nexus is a comprehensive threat intelligence platform powered by AI, machine learning and real-time threat intelligence. The Proofpoint Nexus AI Framework integrates six powerful cores to counter AI-driven threats empowering defenders with advanced tools: Nexus Language Model (LM) combats business email compromise (BEC). It carefully examines email content to detect common elements found in BEC attempts, such as transactional language and urgency. By recognizing subtle linguistic patterns and behavioral cues, Nexus LM for BEC identifies suspicious emails before they can cause harm. Nexus Generative AI automates data analysis across email, cloud and endpoints to identify nuanced patterns in phishing and exfiltration attempts. It also automates complex workflows, isolating compromised devices, revoking access and neutralizing phishing campaigns. Nexus Threat Intelligence (TI) provides real-time updates on attacker tactics, techniques and vulnerabilities, enriching threat detection models. It ensures Proofpoint solutions stay ahead of evolving cyberthreats, offering proactive detection and defense. Nexus Relationship Graph (RG) monitors user behavior across systems, detecting anomalies that signal insider threats or account compromise. By using behavioral analytics, ML and anomaly detection, Nexus RG spots deviations from normal user actions that may indicate a potential threat. Nexus Machine Learning (ML) powers predictive threat detection, which maps known attack behaviors and unsupervised techniques that detect unknown anomalies. It uses behavior-focused detection models to identify malicious activity based on runtime behaviors rather than static signatures. Nexus Computer Vision (CV) is an AI-powered module designed to identify and neutralize vision-based threats. Through advanced computer vision technology, Nexus CV detects threats hidden in visual elements, such as phishing sites, QR codes, malicious attachments and spoofed emails. Good, bad and ugly-AI is complex in the real world AI helps cybersecurity teams as well as the cybercriminals who are trying to outsmart them. What follows are just some of the ways that AI can both help and hinder cybersecurity efforts. The Proofpoint Nexus Framework powers advanced solutions for each of these challenges. The good: AI as a game-changer for cybersecurity In many ways, AI makes it easier for defenders to do their jobs. When it comes to combating sophisticated threats, it can be extremely useful because it addresses the challenges that human teams cannot resolve at scale. Here\'s how it can be taken advantage of when you use Proofpoint. 1: Predict threats before they happen. AI enables a proactive defense to cyber threats. That\'s because AI models can predict potential attack vectors by analyzing historical threat patterns and real-time data. Proofpoint Core Email Protection uses Nexus ML to process billions of data points daily to identify emerging phishing campaigns, malware payloads and zero-day threats. By applying ensemble techniques, Core Email Protection accurately correlates new email-based threats with established patterns. As a result, it helps defenders proactively stop attacks before they infiltrate networks. 2: Detect anomalies acros |
| Notes | ★★ |
| Envoyé | Oui |
| Condensat | accelerates access accidental account accurately across actionable actions activity actor actors adaptive addresses adoption advanced advantage advantages adversaries against ahead ai: alert alerts also altered although analysis analyst analysts analytics analyze analyzed analyzing angle anomalies anomaly answers any apis application applications applying approach are artificial assisted ato attachments attack attacker attackers attacks attempt attempts atypical audio automates automation bad bad: balances based bec because been before behave behavior behavioral behaviors billions block blocked blog both brand browser built business but by ai campaigns can cannot capabilities carefully catch cause certain challenges change changer clear cloud code codes combating combats combined combines come comes common commonly communications complete completely complex complexity comprehensive compromise compromised computer consolidates contact containment content context contextual continuously core cores correlates counter coverage craft create critical cues customers cyber cybercriminal cybercriminals cybersecurity cybersecurity: cyberthreats daily dashboard data day dcg deceive decision decisions deepfake defend defenders defending defense defenses deploy deployment designed designs detailed details detect detecting detecting insider detection detects determine develop deviation deviations devices digital diverse dlp doesn don driven dynamic dynamically each easier easy effectively efforts elements email emails emerging employees empowering enable enables encryption endpoint endpoints enforced engineering enhanced enriching ensemble ensures ensuring entry established evade every evidence evolving examines example excessive executive exfiltration existing expertise explainability explainable explores extremely failed fake false fatigue feeds file flagged flexible focus focused follows fortunately found framework from game generate generative get global good good: governance graph happen harm has help help: helps here hidden high highly hinder hindered historical hours how human identifies identify identifying immense impact impersonated impersonating impersonation inboxes incident incidents includes including indicate indicates infiltrate information infrastructure insider insights integrate integrates integrating integration intelligence interacted intervention irrelevant isolating isolation its jobs just know known language learn learning learning and legitimate linguistic locations login logins logos look looking loss lot machine makes making malicious malware manual many maps may media message metadata methods misaddressed mismatched mitigate mitigates model models module monitors most multiple navigating needed networks neutralize neutralized neutralizing new nexus nexus is nexus power nexus® ai noise normal not now nuanced offering offers often one opaque operational operations organization organizations outsmart overwhelms pages pattern patterns payloads people personalized phishing platform platforms point points policies polymorphic poses positives potential powered powerful powers practical practicality predict predicting predictive preventing malware prioritizes proactive proactively process promise proofpoint proofpoint: protecting protection provides ransomware rather real recognizing redirect reduces relationship rely remediation removing repeated reports requires resolve response result revoking revolutionizing runtime saas safeguard same sandboxing scalable scale scan scripts seconds secure security senior severity sharing sides sifting signal signature signatures simplifies sites six soc social solutions solved some sophisticated speed spoofed spots static stay stop stopped stops structure subtle such suspicious system systems tactics take taken takeover takeovers targeted teams techniques technology than that them there these threat threats threats or through time today tone tool tools: transactional transform trust trying ugly ugly: unauthorized unknown unparalleled |
| Tags | Ransomware Malware Tool Vulnerability Threat Prediction Cloud |
| Stories | |
| Move | 
|
L'article ne semble pas avoir été repris aprés sa publication.
L'article ne semble pas avoir été repris sur un précédent.