One Article Review
| Source |  GoogleSec |
|---|---|
| Identifiant | 8493537 |
| Date de publication | 2024-04-26 18:33:10 (vue: 2024-05-04 12:06:53) |
| Titre | Accélération de la réponse aux incidents en utilisant une AI générative Accelerating incident response using generative AI |
| Texte | Lambert Rosique and Jan Keller, Security Workflow Automation, and Diana Kramer, Alexandra Bowen and Andrew Cho, Privacy and Security Incident Response IntroductionAs security professionals, we\'re constantly looking for ways to reduce risk and improve our workflow\'s efficiency. We\'ve made great strides in using AI to identify malicious content, block threats, and discover and fix vulnerabilities. We also published the Secure AI Framework (SAIF), a conceptual framework for secure AI systems to ensure we are deploying AI in a responsible manner. Today we are highlighting another way we use generative AI to help the defenders gain the advantage: Leveraging LLMs (Large Language Model) to speed-up our security and privacy incidents workflows. |
| Notes | ★★★ |
| Envoyé | Oui |
| Condensat | comparison evolving given input managing our today ui what 200 300 able about above: accelerating accept accountability accountthe accuracy accurate accurately acronyms actions active added additional address adhere adherence advanced advantage: affected after again ai our alert alexandra aliases all allow allowed almost also always ambitious andrew another any apply approach approaches approachwhen appropriate are area areas around assessing audiences author automation based be: because becomes behalf being best beyond billions block both bowen brought build built but button c++ call can capabilities cases cases: cause changes cho closely closing closure closure: code codes/logs commander commanders communication communications complementing completenesscomparison complex complexity compliance component conceptual concise conclude conclusions confidential constantly consuming contain content continuous continuously contractual convey coordination coordination: could covered crafted critical crucial current customers cut cyber data date defenders delivering depends deploying design designed detect detection determined development diagram diana did different digest discard discover documents does draft drafted drafts draw during eager early edge effective efficiency efficient efficiently efforts either elaborate email enabler engineer engineering engineeringonce english ensure ensuring equivalents errors especially estimate events every everyday evolve example examples executive executives expectingat experiment experts explained explanatory explore explored exploring fact factor factors facts factual fake faster feedback field final finest first five fix fixed fixes flowleveraging focus follow following form format forward framework free freeing from gain gathering generate generated generative getting global goal google great group growing growth guidelines had half hallucinations hallucinationsfor hand happen harm has have heavily help high higher highlight highlighting hour hours how human humans hypotheses hypothesized ics identification identify images immediately impact implement implicit important improve improved improvement improvement: improvements improving incident incident quantitative incidents incidentsome include including inconsistent incorrect incredible incremental indication indicators information informs infrastructure initial input inputprompt inserted instead integrated integrating integrationin interface internal introduced irrelevant issue its itself jan just keep keller key kramer lambert landscape language large latest leads learned least lessons let level leveraging like links llm llms llmsgiven logging logs long look looking lost made main maintained maintenance malicious manage management manner manual mechanisms meet meeting memory messy metrics might minimize minor mirrored misinterpretation mitigate mitigation mitigationthe mixture model modern monitor monitoring more moreover most much multiple must native nature nearly necessary needed neutral new newly nextwe noisy non normal not noticed off often ones only order other output outstanding over par part partner parts passive per pioneer pipeline plan points populate possible potential practices pre presented privacy privacyleveraging privileged problem problems process processes processing processingthe produced professionals program program:identification: programmatically: programs projects prompt prompt:the promptthis properly proposed protect protecting proved provide published purpose quality quickly ran rated read recommendations reduce reduced refined regulatory relation relay relevant remediation repeating replaced report reported reports requirements resolution resolution: resolved resources responders response responseintroductionas responsible restore resulting results revealing reviewed reviewing rewrite rigorous risk risks root rosique rust safe safety saif same sample save saved saving savings scale second sections secure security seen self sensitive serves ser |
| Tags | Tool Threat Industrial 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.