OpenAI will watermark ChatGPT outputs by default—but only in the EU

OpenAI will watermark ChatGPT outputs by default—but only in the EU

fivenewscrypto    octobre 08, 2026   

OpenAI will begin automatically watermarking text generated with ChatGPT in the European Union, the company has announced. It will also offer the watermarking feature in other regions, but it will be off by default outside of the EU.

The move in Europe is driven by a need to comply with the EU AI Act, which took effect in August. It requires marking content produced by AI models in a way that another tool can detect. Unfortunately, there is still no completely effective and reliable way to do that. A few standards already exist, like SynthID and the C2PA project, but they are relatively easy to circumvent for anyone with basic know-how.

The same is likely true for OpenAI's watermark. Its method is proprietary; the company calls it textGrain, and has published a technical paper explaining how it works. But in general, it works like other LLM watermarking tools we've seen in the past: It puts patterns in the word choices that are not clear to a human reader and that don't meaningfully change the general quality of the output, but that someone with a key can use a specialized detector to find. OpenAI says it will be giving access to the detector to a limited number of researchers and organizations and providing a request-for-approval process for others to be added over time.

Read full article

Comments

HBO Max, Paramount+, and Discovery+ "will unify into a single service”

HBO Max, Paramount+, and Discovery+ "will unify into a single service”

fivenewscrypto    octobre 08, 2026   

After months of industry and customer speculation, Paramount Skydance has shed some light on the future of the HBO Max and Paramount+ streaming services.

Paramount Skydance completed its acquisition of Warner Bros. Discovery (WBD) today, ending a monthslong saga that began when Netflix last year announced its own plans to buy WBD. Netflix scrapped those plans in February. Skydance is the name of the company created from the $111 billion merger.

In an announcement today, Skydance said that all of the direct-to-consumer streaming services it owns, including the Discovery+ service that Discovery Inc. (which merged with WarnerMedia in 2022 to create WBD) launched in 2021, “will unify into a single service over time.”

Read full article

Comments

It looks like the Atlantic storm season may finally produce a hurricane

It looks like the Atlantic storm season may finally produce a hurricane

fivenewscrypto    octobre 07, 2026   

A low-pressure system in the western Gulf of Mexico should soon become a tropical storm, and based on the latest forecast modeling, it has a decent chance of becoming a hurricane.

Normally, a hurricane forming in early October would not be particularly noteworthy. However, this year the development of a hurricane would be a big deal—because it has not happened yet.

In the modern hurricane record—an era dating to about World War II when there has been aircraft reconnaissance and later continuous satellite coverage—there has never been an Atlantic season without a hurricane, and no season has gone this late without one forming.

Read full article

Comments

Hackers obtain counterfeit TLS certificates for Google and other large services

Hackers obtain counterfeit TLS certificates for Google and other large services

fivenewscrypto    octobre 07, 2026   

Attackers hijacked three top-level domains and used their control to mint counterfeit TLS certificates for Google and other large organizations, Google said Tuesday.

The attackers launched a series of attacks on the .gh, .sl, and .as country code top-level domains (ccTLDs) and then modified authoritative DNS records for selected domains within those namespaces. By controlling those DNS records, the attackers were able to pass automated domain control validation checks and obtain unauthorized certificates for “several Google domains” and “several leading global brands and widely used online services.” Google said it updated Chrome to block all certificates it identified as unauthorized, and worked with the issuing certification authorities to ensure the unauthorized certificates for Google properties were revoked.

Certificate issuance: The weak link in the chain

TLS certificates are the cryptographic credentials that underpin authentication and encryption protections for websites, mail servers, and other Internet infrastructure. These x.509 certificates use a digital signature to bind a domain name such as google.com to a public key. The public key is publicly available, while the private key is held only by the website operator. When a connection shows that the keys match, the visiting party knows it’s connected to the authentic site rather than an impostor. Possession of unauthorized certificates allows attackers to cryptographically impersonate the affected infrastructure.

Read full article

Comments

Paramount completes $111B Warner merger, creating “Skydance” behemoth

Paramount completes $111B Warner merger, creating “Skydance” behemoth

fivenewscrypto    octobre 07, 2026   

Paramount Skydance completed its $111 billion merger with Warner Bros. Discovery today, after a last-ditch effort to block the deal was rejected by Supreme Court Justice Elena Kagan.

The post-merger company is called Skydance, taking the name of a firm that Paramount bought in a separate deal last year. The new Skydance combines two of the largest movie studios, streaming services Paramount+ and HBO Max, CBS, CNN, "and a portfolio of live sports including CBS Sports and TNT Sports, as well as a deep programming library and expansive collection of brands and franchises," Skydance said today in a press release announcing completion of the deal.

The merger was delayed a bit by a lawsuit filed by California and 11 other states. In July, US District Judge Araceli Martínez-Olguín in the Northern District of California ruled that the combination would likely reduce competition substantially and violate antitrust laws.

Read full article

Comments

Amazon kills Alexa’s ability to control Echo speakers' AUX input

Amazon kills Alexa’s ability to control Echo speakers' AUX input

fivenewscrypto    octobre 07, 2026   

Amazon has disabled Alexa control of devices plugged into the company's Echo smart speaker’s AUX port.

Users have reported receiving an email from Amazon this week informing them of the change. “This is a reminder that as of today, voice commands to control your Echo device’s Aux input are no longer available. Requests like 'Alexa, play from line in” or “Alexa, set aux input' will no longer work,” an email that Amazon sent reads, according to a screenshot a Reddit user shared this week.

An Amazon spokesperson confirmed the news to Ars Technica today.

Read full article

Comments

MCP for agent-to-agent comms may be the riskiest protocol you've never heard of

MCP for agent-to-agent comms may be the riskiest protocol you've never heard of

fivenewscrypto    octobre 07, 2026   

The adoption of AI agents in millions of organizations is creating new opportunities for attackers to make them take malicious actions, such as exfiltrating database contents and sensitive business and personal information.

In the past five months, Google and four other organizations—with little in common except for their use of AI agents—have acknowledged vulnerabilities that exploit one agent inside a targeted network to spread harmful instructions to other internal agents. The technique is a special form of prompt injection that targets not the LLM but a particular agent, such as one for translation or data analysis. Guardrails inside such agents, if they exist at all, are often lax and will send the instructions to other agents down the chain. Because the latter agent explicitly trusts the first one, it follows the directions.

Unexpected and hard to mitigate

Independent researcher Syed Anas Mohiuddin tested agents from organizations including Google, JP Morgan Chase, Weviate, Rapid7, the French government's interministerial digital directorate, and the US federal government. His proof-of-concept attacks exploit trust gaps in MCP, short for Model Context Protocol. The standard is one way AI apps and agents communicate with each other inside an internal network. The illustration below shows a simplified MCP in action.

Read full article

Comments