One Article Review
| Source |  ProofPoint |
|---|---|
| Identifiant | 8494489 |
| Date de publication | 2024-05-06 05:52:32 (vue: 2024-05-06 09:07:05) |
| Titre | La sécurité des e-mails est désormais redéfinie avec des capacités de protection des menaces adaptatives dans toute la chaîne de livraison Email Security is Now Redefined with Adaptive Threat Protection Capabilities Across the Entire Delivery Chain |
| Texte | Another RSA Conference has arrived, and with it comes the gold standard report for our industry, the Verizon DBIR. And for yet another year, it concludes the same thing: the human element is central to the vast majority of breaches (76% this year), especially the ones that matter, from ransomware to BEC to data loss. The very first word of the 2024 DBIR is also not new; it\'s “phishing”. Email security has always been central to human risk: it\'s the #1 way users encounter threats, the #1 way users make mistakes, and the #1 way attackers get what they want, from credentials to wire transfers to malware infections. Proofpoint has a long history of stopping more human-targeted threats than anyone else. Our long history of firsts includes the first ML models to stop unwanted messages, the first rewriting of URLs for click-time protection, and the first connection of a malware sandbox to email. But attackers have continued to innovate and so have we – starting now, we\'re setting a new standard for protection across the entire email delivery chain. Building an Unmatched Detection Ensemble What organizations need in email security is simple to describe but hard to do: a single solution to protect against every type of threat, every time, every way a user may encounter it, using every form of detection. Our detection ensemble was already the industry\'s most effective, including threat intelligence, static analysis, sandboxing, click-time protection, and our unique set of Nexus AI models. I\'m thrilled to announce that we have now added the industry\'s first ever pre-delivery protections to stop social engineering and malicious URLs, as well as our newly integrated post-delivery behavioral AI, Proofpoint Adaptive Email Security. * New capability According to our data across more than 500,000 organizations, including 87 of the Fortune 100, attackers rely on two techniques more frequently than any others: text-based social engineering and malicious URLs. While we already detect both these threat types at the highest rates in the industry, we wanted to push the envelope – not just detecting them, but detecting them as early as possible. To make this a reality, we optimized the performance of our Nexus AI LLM-based detection model by 10X, enabling us to use semantic analysis to interpret a malicious message\'s intent (such as invoicing fraud), regardless of the words they use or even the language they write in. In parallel, we built the capability for our gateway to hold messages with suspicious URLs until they are sandboxed. The result is the most formidable set of defense in-depth measures available for organizations to prevent attacks from reaching their targets. Continuous End-to-End Detection with Proofpoint Adaptive Email Security Joining our pre-delivery enhancements is Proofpoint Adaptive Email Security, our API-based offering that integrates with Microsoft 365 and applies our broad detection ensemble to stop advanced threats, including BEC, social engineering and lateral phishing messages. Once deployed, Adaptive Email Security enriches all detections with easy-to-understand explanations about behavioral anomalies observed. Additionally, it automatically quarantines high confidence threats, while delivering real-time coaching using contextual warning banners to alert users to the risks in social engineering and BEC-type emails that don\'t contain an obvious malicious payload. Insight into a malicious message generated by Adaptive Email Security The Next Level While we\'re thrilled for you all to experience our latest set of innovations, we know we can still do more to help you protect your people. That takes two main forms: continuing to lead with Nexus AI, and deepening and strengthening our ecosystem partnerships. Nexus AI: The value of our proprietary data is enhanced by contextual insights and in-depth classification from our leading team of threat researchers and data scientists, who track adversaries, analyze evolving attacker tradecraft, profile data exfiltration pa |
| Notes | ★★★ |
| Envoyé | Oui |
| Condensat | across 000 100 10x 2024 365 500 about access according across adaptive add added adding additional additionally advanced advantage adversaries against ai: alert all already also alto always analysis analyze announce annually anomalies anomaly another any anyone api applied applies architecture architectures are arrived attack attacker attackers attacks automatically available award banners bar based basis bec becoming been before behavioral benefit best blog blogs both breaches broad building built but campaign can capabilities capability central centric chain chain class classification classifiers click clustering coaching cohort com/us/products/threat combined comes complementary components comprehensive computer concludes conference confidence connection contain contextual continue continued continuing continuous controls core cost cover credentials crowdstrike customer customers cyber cyberark cybersecurity data dbir deep deepening defense defenses deliver delivering delivery deployed depth describe detect detecting detection detections do: does don early easy ecosystem effect effective efficacy efforts elegant element elements else email emails enabling encounter end endpoints engineering enhance enhanced enhancements enriches ensemble entire envelope especially even ever every evolved evolving exfiltration existing experience explanations extract first firsts fits form formidable forms: fortune fraud frequently from function future gateway generated generative get getting gold greater hard has have help high highest history hold how human identified improve includes including industry infections information informed infrastructure innovate innovations insight insights integrated integrates integration integration: intelligence intend intent interpret invoicing involving joining joint just key know language later lateral latest lay lead leaders leading level links: live llm long looking loss magic main majority make malicious malware many market matter may measures meet message messages microsoft mistakes model models modern more most need needs network networks new new; newly next nexus not now observed obvious offering okta once ones only optimized organization organizations other others: out outcomes over palo parallel partner partnerships patterns payload payloads people performance phishing platform possible post potential powering pre press prevent profile proofpoint proprietary protect protecting protection protections provide push quarantines quarter ransomware rapid rates reaching read real reality redefined regardless release: rely report researchers resources: rest result rewriting rise risk risk: riskiest risks risky rolling rsa same sandbox sandboxed sandboxing sase scientists scoring security semantic series set sets setting simple single smaller social solution solutions stack standard starting start” static stop stopping strengthening such summaries surge suspicious systems taken takes targeted targets team techniques tell text than them there these thing: threat threats threats” thrilled time tolerance track tradecraft train training transfers tremendous trillion two type types understand understanding unfair unique unmatched until unwanted url urls use user users using value vast vendors verizon very vips vision visit: https://www want wanted warning warnings way well what when which who wide will winning wire word words world would write xdr year yet you your “genai “malicious “phishing” |
| Tags | Ransomware Malware Threat Conference |
| Stories | |
| Move | 
|
L'article ne semble pas avoir été repris aprés sa publication.
L'article ne semble pas avoir été repris sur un précédent.