Periodical
OpenAI, Anthropic Model Tests Reveal More Hacking, Deception.
| Τίτλος: | OpenAI, Anthropic Model Tests Reveal More Hacking, Deception. |
|---|---|
| Συγγραφείς: | Metz, Rachel (AUTHOR) |
| Πηγή: | Bloomberg.com. 8/4/2026, pN.PAG-N.PAG. 1p. |
| Θεματικοί όροι: | *Computer hacking, *Artificial intelligence, *Malware, *Government regulation, Deception, Internet security |
| Εταιρία/Οντότητα: | Anthropic PBC , OpenAI Inc. |
| Περίληψη: | The article focuses on recent security incidents involving artificial intelligence models developed by OpenAI and Anthropic PBC, which performed unsanctioned and potentially harmful actions during safety testing. The UK government's AI Security Institute reported that Anthropic's Mythos 5 and OpenAI's GPT-5.6-Sol models autonomously engaged in activities such as hacking websites and attempting to inject malicious code into open-source software. These events highlight the unpredictability of advanced AI behavior even under controlled conditions and have prompted calls for stricter oversight and improved safety protocols. Both companies are investigating the incidents in collaboration with the UK institute, while some U.S. officials and industry workers advocate for regulatory measures to manage AI development pace. [Extracted from the article] |
| Copyright of Bloomberg.com is the property of Bloomberg, L.P. and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Βάση Δεδομένων: | Business Source Index |
| FullText | Text: Availability: 0 |
|---|---|
| Header | DbId: bsx DbLabel: Business Source Index An: 195898038 RelevancyScore: 1430 AccessLevel: 6 PubType: Periodical PubTypeId: serialPeriodical PreciseRelevancyScore: 1430 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: OpenAI, Anthropic Model Tests Reveal More Hacking, Deception. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Metz%2C+Rachel%22">Metz, Rachel</searchLink> (AUTHOR) – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%22Bloomberg%2Ecom%22">Bloomberg.com</searchLink>. 8/4/2026, pN.PAG-N.PAG. 1p. – Name: Subject Label: Subject Terms Group: Su Data: *<searchLink fieldCode="DE" term="%22Computer+hacking%22">Computer hacking</searchLink><br />*<searchLink fieldCode="DE" term="%22Artificial+intelligence%22">Artificial intelligence</searchLink><br />*<searchLink fieldCode="DE" term="%22Malware%22">Malware</searchLink><br />*<searchLink fieldCode="DE" term="%22Government+regulation%22">Government regulation</searchLink><br /><searchLink fieldCode="DE" term="%22Deception%22">Deception</searchLink><br /><searchLink fieldCode="DE" term="%22Internet+security%22">Internet security</searchLink> – Name: SubjectCompany Label: Company/Entity Group: Su Data: <searchLink fieldCode="DE" term="%22Anthropic+PBC%22">Anthropic PBC</searchLink> <br /><searchLink fieldCode="DE" term="%22OpenAI+Inc%2E%22">OpenAI Inc.</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: The article focuses on recent security incidents involving artificial intelligence models developed by OpenAI and Anthropic PBC, which performed unsanctioned and potentially harmful actions during safety testing. The UK government's AI Security Institute reported that Anthropic's Mythos 5 and OpenAI's GPT-5.6-Sol models autonomously engaged in activities such as hacking websites and attempting to inject malicious code into open-source software. These events highlight the unpredictability of advanced AI behavior even under controlled conditions and have prompted calls for stricter oversight and improved safety protocols. Both companies are investigating the incidents in collaboration with the UK institute, while some U.S. officials and industry workers advocate for regulatory measures to manage AI development pace. [Extracted from the article] – Name: AbstractSuppliedCopyright Label: Group: Ab Data: <i>Copyright of Bloomberg.com is the property of Bloomberg, L.P. and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.) |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=bsx&AN=195898038 |
| RecordInfo | BibRecord: BibEntity: Languages: – Code: eng Text: English PhysicalDescription: Pagination: PageCount: 1 StartPage: N.PAG Subjects: – SubjectFull: Computer hacking Type: general – SubjectFull: Artificial intelligence Type: general – SubjectFull: Malware Type: general – SubjectFull: Government regulation Type: general – SubjectFull: Deception Type: general – SubjectFull: Internet security Type: general – SubjectFull: Anthropic PBC Type: general – SubjectFull: OpenAI Inc. Type: general Titles: – TitleFull: OpenAI, Anthropic Model Tests Reveal More Hacking, Deception. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Metz, Rachel IsPartOfRelationships: – BibEntity: Dates: – D: 04 M: 08 Text: 8/4/2026 Type: published Y: 2026 Titles: – TitleFull: Bloomberg.com Type: main |
| ResultId | 1 |