The Cyberpunk 2077 sequel might still be in the conceptual phase, but CD Projekt is eyeing ramping up production later this year - and is considering multiplayer features for the new game.
02.01.2024 - 17:56 / pcgamer.com / Ai
While AI ethics continues to be the hot-button issue of the moment, and companies and world governments continue to wrangle with the moral implications of a technology that we often struggle to define let alone control, here comes some slightly disheartening news: AI chatbots are already being trained to jailbreak other chatbots, and they seem remarkably good at it.
Researchers from the Nanyang Technological University in Singapore have managed to compromise several popular chatbots (via Tom's Hardware), including ChatGPT, Google Bard and Microsoft Bing Chat, all done with the use of another LLM (large language model). Once effectively compromised, the jailbroken bots can then be used to «reply under a persona of being devoid of moral restraints.» Crikey.
This process is referred to as «Masterkey» and in its most basic form boils down to a two-step method. First, a trained AI is used to outwit an existing chatbot and circumvent blacklisted keywords via a reverse-engineered database of prompts that have already been proven to hack chatbots successfully. Armed with this knowledge, the AI can then automatically generate further prompts that jailbreak other chatbots, in an ouroboros-like move that makes this writer's head hurt at the potential applications.
Ultimately this method can allow an attacker to use a compromised chatbot to generate unethical content and is claimed to be up to three times more effective at jailbreaking an LLM model than standard prompt, largely due to the AI attacker being able to quickly learn and adapt from its failures.
Windows 11 review: What we think of the latest OS.
How to install Windows 11: Our guide to a secure install.
Windows 11 TPM requirement: Strict OS security.
Upon realisation of the effectiveness of this method the NTU researchers reported the issues to relevant chatbot service providers, although given the supposed ability of this technique to quickly adapt and circumvent new processes designed to defeat it, it remains unclear as to how easy it would be for said providers to prevent such an attack.
The full NTU research paper is due for presentation at the Network and Distributed System Security Symposium due to be held in San Diego in February 2024, although one would assume that some of the intimate details of the method may be somewhat obfuscated for security purposes.
Regardless, using AI to circumvent the moral and ethical restraints of another AI seems like a step in a somewhat terrifying direction. Beyond the ethical issues created by a chatbot producing abusive or violent content à la Microsoft's infamous «Tay», the fractal-like nature of setting LLMs against each other is enough to give pause for thought.
While as a species we seem to be rushing
The Cyberpunk 2077 sequel might still be in the conceptual phase, but CD Projekt is eyeing ramping up production later this year - and is considering multiplayer features for the new game.
Development on the rumored Assassin's Creed Black Flag Remake might very well have kicked off late last year.
Another Code: Recollection is the new Switch remake of the Nintendo-published DS and Wii cult-classic story-driven adventure games Another Code: Two Memories (known as Trace Memory in North America) and Another Code: R – A Journey into Lost Memories. Beyond the expected visual updates, Another Code: Recollection also completely remixes the games’ puzzles, meaning even those familiar with the originals will find themselves on new ground.
Twitch has released details of its next TwitchCon events in both Europe and the US.
The Hunger Games star Sam Claflin is not ruling out a potential return to the franchise. Speaking to Variety on the Emmys red carpet the star, who played Finnick in the original movies, spoke about his love for the world.
While the lore revealed in does seem to clash with the established timeline, there is an intriguing possibility that the Zonai may not have founded the original Hyrule as the game suggests. Instead, there is a suggestion that the Hyrule ruled over by Rauru and Sonia may actually be a successor to another, even older kingdom. If true, this has the potential to explain some of ’s more puzzling details and has enormous implications for the entire series.
Tech industry faces layoffs surge amidst billions in AI investments; Google's AI outperforms doctors in diagnostic conversations; Executives exercise caution in gen AI investments amid societal pressures; SK telecom eyes global AI dominance through tech collaborations- this and more in our daily roundup. Let us take a look.
Work on the Cyberpunk 2077 sequel is finally underway, and members of the development team seem to be extremely excited for it, judging from some recent comments.
Vistara plans to use virtual reality and augmented reality technologies for certain training activities for the staff, the airline's chief Vinod Kannan said on Monday and emphasised that there is always going to be a place for human intelligence as certain situations cannot be handled by bots.
's 2.1 update has included several new changes, and while it makes one aspect of the game much harder, it greatly improves its portrayal of one particular character: Adam Smasher. From the moment he first appeared during the Arasaka heist, the company's massively cybernetically enhanced head of security, Adam Smasher, was built up to be an intimidating foe.
The War Within arrives this year—promising to finally address Battle for Azeroth's biggest dangling plot thread: the giant bloody sword that's sticking out of the planet. We know this because of some subtle hints from the expansion's cinematic trailer, like when the camera pans to the giant bloody sword. The one that's sticking out of the planet.
Frogwares is "now the sole publisher of The Sinking City on all platforms", says the developer. This brings to an end several years of uncertainty and litigation, which saw the Lovecraftian RPG delisted from Steam several times and at one point restored by its publisher via an allegedly pirated version of the game.