Accueil > Accueil > Global Security Mag Online
http://www.globalsecuritymag.fr/
Articles
-
Multiples vulnérabilités dans Oracle Weblogic (16 septembre 2026)
16 septembreDe multiples vulnérabilités ont été découvertes dans Oracle Weblogic. Elles permettent à un attaquant de provoquer une exécution de code arbitraire à distance.
Voir en ligne : https://www.cert.ssi.gouv.fr/avis/C... -
Palantir, Nvidia and Booz Allen restrict AI models over data security concerns
16 septembre, par Black Hills Information Security, Suzu Labs, Xcape Inc. and Ridge Security — afficheMajor technology and defense companies including Palantir, Nvidia and Booz Allen Hamilton are restricting or considering ending their use of advanced AI models unless providers including Anthropic and OpenAI offer stronger guarantees that sensitive corporate data and intellectual property will not be retained or misused.
Palantir has pressed Anthropic for irrevocable zero-data-retention guarantees before offering its latest model through Palantir's software. Nvidia is limiting Anthropic's models to less-sensitive work and using its own Nemotron models for proprietary projects, while Booz Allen has barred employees from using Anthropic's commercial model for cybersecurity work involving proprietary software.
In June, Anthropic changed its data policy to retain usage logs for 30 days to help detect complex attacks spanning multiple sessions. Anthropic and OpenAI both say they do not train their models on customer data by default unless customers opt in.
Experts with Black Hills Information Security, Suzu Labs, Xcape Inc. and Ridge Security offer perspectives on the matter.
John Strand, Owner, Black Hills Information Security, Inc.:
“This is kind of insane, and it reminds me a lot of what happened with cloud computing years ago. When companies started moving to Microsoft and Google cloud services, there was an outcry. People were saying, ‘Our data isn't on our servers anymore. It's on Google's servers.' Eventually, that died down. People got used to the reality that their data was going to be stored in the cloud.
“I think AI is having a very similar moment. People are looking at Anthropic and OpenAI and getting upset about 30-day log retention policies. You can scream and thrash about that all you want, but the reality is that logging is a fundamental part of security and data governance. We keep transactional database logs. We keep email and activity logs in systems like Microsoft Purview. Organizations have requirements to maintain these records for security, auditing, governance, and accountability.
“When I look at Anthropic's response in particular, it feels like they're trying very hard to make sure these transactional logs exist for security and accountability.
“And if your data is so sensitive that even that level of retention is unacceptable, there is another option. Run open-weight models locally on infrastructure you control. That's exactly what we do for particularly sensitive activities.
“Logging and analysis have always been critical parts of security. AI doesn't suddenly change that.”
Steven Swift, Managing Director, Suzu Labs:
“Organizations who prefer the old zero retention guarantees are likely to get stuck using older models. Anthropic and other labs have very publicly been fighting against wide scale misuse of their products. In order to meaningfully find and fix novel abuse techniques, it is essential that Anthropic maintain logs for a period of time. Its basically not possible to build adequate safeguards without such logs.
“I would expect that as time goes on, that all newer models carry with them the same standard data retention requirement. As time goes on, organizations will slowly get used to the 30 day retention window. It seems likely we'll see the 30 day retention expanded to include older models as well. Especially where some of the older models clearly don't have the same sorts of safeguards in place, making abuse on these older models more prevalent.
“There isn't much upside for any of the frontier labs to offer true zero retention usage of their services. They already offer a guarantee that the data retained during the 30 day window won't be used for model training.”
Donald McFarlane, Board Member, Xcape Inc.:
“We need to stop treating digital information as something people or businesses surrender simply because they place it in someone else's custody. We should treat that relationship for what it is: a digital bailment.”
“Possession is not ownership. When I entrust proprietary code, business records, personal information, or other unpublished material to a cloud or AI provider, that must be deemed to create duties much like a traditional bailment: use it for the agreed purpose, protect it, and do not retain, repurpose, sell, or disclose it beyond that purpose.”
“AI makes this especially important because aggregation magnifies the risk. Thousands of individually mundane interactions can collectively reveal extraordinarily sensitive information about a person or company.”
“Moving our papers from a desk drawer or filing cabinet into the cloud, including into the memory and logs of an AI model harness, should not strip away our property rights or make it easier for government to obtain them without proper legal process.”
Claudionor Coelho, Senior Advisor, Ridge Security:
“There is a discussion on whether OpenAI used data or meta-data to accelerate and anticipate the proof of Navier Stokes equations.”
“In addition to that, Anthropic announced recently it is opening several verticals (like biotech, finance, etc). That raises the questions on whether any company's data or meta-data are being used by the foundation labs, if not to train their models, at least using meta-data to help make decisions on which verticals to enter.”
-
Cyberattack knocks International Meteor Org offline for weeks
15 septembre, par Black Hills Information Security — afficheThe International Meteor Organization (IMO), a Belgium-based nonprofit that coordinates meteor observations worldwide, says a cyberattack dealt a “critical blow” to its aging IT infrastructure, taking much of its website and online services offline. The organization expects several weeks of partial downtime while it moves to new infrastructure and services.
IMO has restored its fireball reporting system, which allows people around the world to submit observations of unusually bright meteors, but other parts of its website remain unavailable.The organization has not disclosed when the attack occurred, how attackers gained access or whether any information was accessed, stolen or encrypted. The incident has forced IMO to rebuild portions of its infrastructure rather than simply restore its website.
IMO has also not said whether information belonging to its members, meteor observers or other contributors was exposed.An expert with Black Hills Information Security offers perspective on the matter.
Bronwen Aker, Report Editor, Black Hills Information Security:
“An organization that tracks space debris burning up in the atmosphere for free doesn't seem like a juicy target for malicious hackers, so an attack against the International Meteor Organization makes little obvious sense from a financial perspective. It doesn't make much more sense if the goal was to harvest credentials for identity theft or other forms of fraud, either."
“Without more information, there is no way to know exactly what happened. And if IMO's infrastructure was as old and fragile as the organization suggests, they may never know. Older systems are not known for comprehensive logging, and logging is one of the things incident responders depend on to reconstruct an attack: how someone got in, what they touched, what they changed, and whether they took anything with them."
“That matters because IMO has not said how the attackers gained access or whether information was accessed, stolen, or encrypted. The absence of an answer does not necessarily mean the organization is withholding one. It may simply mean the evidence needed to answer those questions was never recorded."
“The bigger question for me is: why pick on a bunch of science geeks?"
“Organizations like IMO operate on limited budgets and a great deal of volunteer effort to collect scientific observations and make that information useful to everyone. Maybe the attackers were simply scanning the Internet for vulnerable systems and happened to find one. Maybe IMO was deliberately targeted. We do not know."
“But from the outside, it has the feel of a bully knocking over the nerd simply because he can. Whatever the motive, the result is weeks of recovery work for an organization whose primary mission is watching the sky and sharing what it learns.”
-
Tenable reste n°1 mondial de la gestion des vulnérabilités et de l’exposition selon IDC
15 septembre, par Valentin Jangwa, Global Security Mag — afficheTenable se classe n°1 de la gestion des vulnérabilités et de l'exposition des équipements pour la huitième année consécutive
Tenable attribue cette position à l'évolution continue de sa plateforme, notamment à ses avancées en matière d'IA agentique et à la demande croissante pour la sécurisation des environnements IA.Tenable Holdings, Inc. (NASDAQ : TENB), la société de gestion de l'exposition au risque cyber, annonce avoir obtenu la première place mondiale en part de marché dans le rapport IDC Worldwide Device Vulnerability and Exposure Management Market Shares (document n°US53903826, août 2026).
Avec le développement de l'IA, la gestion de l'exposition constitue une évolution fondamentale et nécessaire de la manière dont les entreprises appréhendent et réduisent les risques. Les équipes de sécurité sont arrivées à un stade où le volume des menaces critiques et complexes a augmenté de façon exponentielle, et elles disposent de moins en moins de temps pour les évaluer et y répondre avant que des attaquants utilisant l'IA ne passent à l'action. La plateforme de gestion de l'exposition Tenable One a été conçue pour répondre à cet enjeu en aidant les entreprises à mieux comprendre leurs cyber-risques, à hiérarchiser leurs actions et à anticiper les attaques.
Selon le rapport d'IDC sur les parts de marché, « Tenable a maintenu sa position de leader incontesté, avec 24,6 % du marché mondial de la gestion des vulnérabilités et de l'exposition des équipements, soit près de 260 millions de dollars d'avance sur son concurrent le plus proche ». Tenable attribue cette position aux évolutions apportées à sa plateforme afin d'unifier la visibilité, de fournir une analyse précise et contextualisée de l'exposition, ainsi que d'accélérer et d'automatiser des actions dans l'ensemble de l'entreprise. Ces derniers mois, Tenable a introduit des capacités autonomes fonctionnant en continu dans Tenable Hexa AI, le moteur d'IA agentique de Tenable One. L'entreprise a également étendu les fonctionnalités de Tenable One AI Exposure à l'ensemble des principales plateformes d'IA et aux outils de développement, élargi sa couverture aux environnements de sécurité applicative et étendu les capacités de contrôle et de validation continus de la sécurité de sa plateforme.
Tenable s'engage également à contribuer à l'évolution de la gestion de l'exposition pilotée par l'IA grâce à l'innovation continue de sa plateforme et à ses collaborations stratégiques avec OpenAI et Anthropic. L'entreprise participe à l'OpenAI Daybreak Defense Network ainsi qu'au projet Glasswing d'Anthropic. Tenable a par ailleurs récemment annoncé la disponibilité de Claude Mythos 5 dans Tenable One afin d'aider les entreprises à mieux comprendre leurs cyber-risques, à hiérarchiser leurs actions et à anticiper les attaques.
« Nous estimons que nos huit années consécutives au premier rang en matière de parts de marché reflètent notre engagement à proposer une innovation continue qui fait référence en matière de gestion de l'exposition », déclare Mark Thurmond, co-CEO de Tenable. « Nous permettons aux entreprises d'adapter leur dispositif de défense aux enjeux de l'IA, en leur apportant une visibilité complète, les analyses avancées et les technologies d'intelligence artificielle nécessaires pour garder une longueur d'avance sur les menaces actuelles. »
« On ne saurait trop insister sur ce point : la gestion de l'exposition constitue un socle indispensable pour comprendre, valider et hiérarchiser les expositions avant qu'elles ne soient exploitées par des attaquants utilisant l'IA », déclare Michelle Abraham, vice-présidente de la recherche, Sécurité et confiance chez IDC. « Les conseils d'administration, les assureurs et les régulateurs incitent les équipes de sécurité à dépasser le simple décompte des vulnérabilités pour évaluer le niveau de risque résiduel. Les organisations qui adoptent les principes de la gestion de l'exposition peuvent réduire les risques, améliorer l'efficacité des mesures correctives et démontrer des progrès mesurables à leurs parties prenantes. »
Plus d'informations sur la plateforme de gestion de l'exposition Tenable One sont disponibles au lien suivant. https://fr.tenable.com/products/tenable-oneÀ propos de Tenable
Tenable est la société de gestion de l'exposition au cyber-risque, qui révèle et comble les failles qui affectent la valeur d'une entreprise, compromettant sa réputation et diminuant la confiance de ses clients. Sa plateforme de gestion de l'exposition optimisée par l'IA unit de manière inédite la visibilité, les analyses et les actions de sécurité sur l'ensemble de la surface d'attaque. Les entreprises sont ainsi équipées pour se défendre contre tout type d'attaque, de l'infrastructure IT aux environnements cloud en passant par les infrastructures critiques. En protégeant les entreprises de l'exposition cyber, Tenable réduit les risques business de plus de 40 000 clients à travers le monde. Pour en savoir plus : fr.tenable.com. https://fr.tenable.com/
GS Days