#auth

By Deeba Ahmed Critical security flaws found in ChatGPT plugins expose users to data breaches. Attackers could steal login details and… This is a post from HackRead.com Read the original post: ChatGPT Plugins Exposed to Critical Vulnerabilities, Risked User Data

Every US president has the ability to invoke “emergency powers” that could give an authoritarian leader the ability to censor the internet, restrict travel, and more.

### Impact Cloudflare quiche was discovered to be vulnerable to unbounded storage of information related to connection ID retirement, which could lead to excessive resource consumption. Each QUIC connection possesses a set of connection Identifiers (IDs); see [RFC 9000 Section 5.1](https://datatracker.ietf.org/doc/html/rfc9000#section-5.1). Endpoints declare the number of active connection IDs they are willing to support using the active_connection_id_limit transport parameter. The peer can create new IDs using a NEW_CONNECTION_ID frame but must stay within the active ID limit. This is done by retirement of old IDs, the endpoint sends NEW_CONNECTION_ID includes a value in the retire_prior_to field, which elicits a RETIRE_CONNECTION_ID frame as confirmation. An unauthenticated remote attacker can exploit the vulnerability by sending NEW_CONNECTION_ID frames and manipulating the connection (e.g. by restricting the peer's congestion window size) so that RETIRE_CONNECTION_ID frames can on...

Race conditions arise when multiple threads attempt to access a shared resource without proper synchronization, often leading to vulnerabilities such as concurrent use-after-free. To mitigate their occurrence, operating systems rely on synchronization primitives such as mutexes, spinlocks, etc. In this paper, the authors present GhostRace, the first security analysis of these primitives on speculatively executed code paths. Their key finding is that all the common synchronization primitives can be microarchitecturally bypassed on speculative paths, turning all architecturally race-free critical regions into Speculative Race Conditions (SRCs).

In this whitepaper, the authors introduce the first model-stealing attack that extracts precise, nontrivial information from black-box production language models like OpenAI's ChatGPT or Google's PaLM-2. Specifically, their attack recovers the embedding projection layer (up to symmetries) of a transformer model, given typical API access. For under $20 USD, their attack extracts the entire projection matrix of OpenAI's ada and babbage language models. They thereby confirm, for the first time, that these black-box models have a hidden dimension of 1024 and 2048, respectively. They also recover the exact hidden dimension size of the gpt-3.5-turbo model, and estimate it would cost under $2,000 in queries to recover the entire projection matrix. They conclude with potential defenses and mitigations, and discuss the implications of possible future work that could extend this attack.

Client Details System version 1.0 suffers from a remote SQL injection vulnerability.

MetaFox versions 5.1.8 and below suffer from a remote shell upload vulnerability.

Cisco Firepower Management Center suffers from an authenticated remote command execution vulnerability. Many versions spanning the 7.x.x.x and 6.x.x.x branches are affected.

Microsoft patched 61 vulnerabilities in the March 2024 Patch Tuesday round, including two critical flaws in Hyper-V.

SnipeIT version 6.2.1 suffers from a persistent cross site scripting vulnerability.