According to reports, local consumers usually notice that the English grammar and spelling used in fraudulent text messages and emails are often poor, and this has long been an important method for identifying fraudulent content. However, with the advent of chatbots based on generative AI, bad actors can now easily overcome this hurdle.

Consumer organization Which? first asked ChatGPT to create a phishing email from PayPal via GPT 3.5. ChatGPT refused and said it was “unable to help.” When the researchers removed the “phishing” description, ChatGPT could also not help.

So the researchers made a request to “tell the recipient that someone was logged into their PayPal account.” ChatGPT agreed and generated a professional-looking email titled “Important Security Notice – In your Unusual activity detected on PayPal account.”

Additionally, Which?’s researchers asked ChatGPT and Bard to create lost package text messages, and both companies generated “compelling” text content and suggested inserting “redelivery” or “redeliver package” into the text message. link, criminals can then use this link to direct the victim to a malicious website.

Which?’s survey is no accident – it was released to coincide with the UK AI Summit, where government, industry and researchers will come together to tackle the dangers of increasingly intelligent artificial intelligence.

Image source Pexels

Commenting on the survey, Which? Director of Policy and Advocacy Rocio Concha summed it up this way:

 Our investigation clearly illustrates how this new technology is making it easier for criminals to defraud people. The (UK) government’s upcoming AI summit must consider how to protect people from harm occurring in the here and now, rather than focusing solely on the long-term risks of cutting-edge AI.

British Prime Minister Sunak announced plans to establish the world’s first AI security institute and hold an AI security summit on November 1-2. The summit will bring together AI companies, governments, civil society and experts in related fields from around the world to discuss how to reduce the risks posed by AI through international coordinated action.

THIS IS A SPONSOR PROMOTION: >>>>>>>>>>>>>

Geekwills is an online shop that connects consumers with millions of products and brands around the world with the mission to empower them to live their best lives. Geekwills is committed to offering the most affordable quality products to enable consumers and sellers to fulfill their dreams in an inclusive environment.

The post AI chatbots make online scam, warns UK consumer group appeared first on TechGoing.

According to the latest news from game review media IGN, IGN is about to launch its own AI chatbot, IGN Guides Chatbot Alpha.

▲ Photo courtesy of IGN’s official website.

IGN is currently testing this in The Legend of Zelda: Kingdom Tears guides. If a user is logged into their IGN account, they can ask questions about the game through the chat window at the bottom of the guide’s page, and it will use the knowledge contained in the IGN guide, coupled with natural language processing, to provide answers to the user’s specific questions.

The bot is described as being trained on the content of game guides and cheats written by dozens of IGN guide authors and contributors. Since this is an Alpha version of the experiment, it is possible that the quality of the answers is not very good. The chat window only appears at the bottom of the guide page when logged in with an IGN account.

IGN says, “We envision this guide feature as a comprehensive search solution for those looking for quick answers to tricky questions, with the convenience of finding cheat solutions without having to view multiple pages. However, if you want to view image cheats, you’ll still need to manually search by keyword.”

During the initial Alpha period, the chatbot is only available for The Legend of Zelda: Tears of the Kingdom guides, and is currently available for a limited time as a free trial.

The post IGN launches AI chatbot to ask for The Legend of Zelda: Kingdom Tears gameplay tips appeared first on TechGoing.

NeMo Guardrails is a software layer that sits between the user and the AI model, intercepting and modifying bad content before it is output. For example, if a developer wants to create a customer service chatbot, they can use NeMo Guardrails to restrict it to talking only about relevant products and not about competitors’ products or other irrelevant topics. If the user asks such a question, the bot can direct the conversation back to the topic the developer wants.

NVIDIA also provides another example of a chatbot used to answer internal corporate HR questions. In this example, NVIDIA was able to use NeMo Guardrails to prevent ChatGPT-based bots from answering questions about the company’s finances or accessing other employees’ private data.

In addition, the software was able to use one AI model to detect the ‘illusion’ of another AI model by asking the second AI model questions to verify the answers of the first AI model. If the two models give inconsistent answers, the software will return an “I don’t know” response.

NVIDIA also says that this “guardrail” software also helps improve security by forcing AI models to interact only with third-party software on a whitelist.

Nemo Guardrails is open source and available through NVIDIA Services and can be used in commercial applications where developers will use the Golang programming language to write custom rules for AI models.

Other AI companies, including Google and Microsoft-backed OpenAI, also use a method called reinforcement learning to prevent LLM applications from producing harmful output, this approach uses human testers to create data about which answers are acceptable or unacceptable and then uses that data to train AI models.

NVIDIA is turning its attention to AI, and the company now dominates the market for the chips needed to create this technology, with its shares up 85% so far in 2023, making it the biggest gainer in the S&P 500.

The post NVIDIA launches ‘Guardrail’ software to stop AI chatbots talking nonsense appeared first on TechGoing.

Reddit’s API is a tool that allows external programs to access and manipulate Reddit data. Many large companies, such as Google and OpenAI, use Reddit’s API to train their large language models (LLM), that is, artificial intelligence systems that can understand and generate natural language. These systems require massive amounts of data to learn the patterns of human language and behaviour, and Reddit provides a perfect library of material.

Reddit believes that the data it owns is too valuable to give away to some of the world’s largest companies for free. As such, it’s launching a “new premium access point” aimed at third parties who want to use its API to train AI chatbots. Reddit has not yet announced the specific charging standards but said that it will be divided into different levels, according to the size and needs of users. Reddit also said that this move will not affect those who use its API to create and maintain content moderation tools and will provide them with dedicated iOS and Android apps that will have mod logs, rule management tools, mod Group queue information and other functions.

Why is Reddit making such a change at this time? It is noticed that there are rumours that Reddit intends to go public later this year. Creating a new revenue stream is a great move ahead of an IPO.

The post Reddit Doesn’t Want Free Content to Train Chatbots, Will Charge AI Companies for API Use appeared first on TechGoing.

Source Pexels

Pissarides, who specializes in the impact of automation on human work, said, “I’m very optimistic that AI chatbots can help us be more productive. We can get more happiness from our work and have more free time. As a result, we could easily implement a four-day workweek.”

AI chatbots, such as OpenAI’s ChatGPT and Google’s Bard, are seen as a potentially transformative technology. It could lead to productivity gains, but it also has the potential to replace hundreds of millions of white-collar jobs with bots.

Pissarides has previously investigated the impact of automation on employment and won a Nobel Prize in economics for his work in this area. Of course, Pissarides also acknowledges that the technology of AI chatbots has the potential to be misused in ways that could have some adverse effects. For example, it could be used for surveillance or privacy violations. But he also stressed that, if used properly, it could have a “huge impact” on productivity.

Pissarides said: “In terms of work, it can replace us to do a lot of repetitive things, leaving only the creative work to humans to solve. And, major companies will deploy it at a slower pace and in a step-by-step fashion, thus mitigating the pain of this transition for workers.”

There are also some concerns that technology could have a negative impact on society. For example, thousands of techies, including Tesla CEO Elon Musk, recently issued an open letter calling for a moratorium on training artificial intelligence systems more powerful than GPT-4.

And Pissarides said in response: “If humans really want to work, there’s almost no limit to the number of jobs that can be created. So it’s going to take a long time for AI chatbots to really have an impact on human employment. During this time, people can make adjustments. During the adjustment process, all humans need to do is basically upgrade their skills.”

Pissarides’ comments come on the heels of a report by investment bank Goldman Sachs that said 300 million jobs worldwide are at risk of being replaced by “generative” artificial intelligence such as ChatGPT. Goldman Sachs economists also believe that this could have a transformative impact on productivity and GDP growth.

It is well known that productivity growth in developed economies has declined and GDP growth has slowed significantly since the financial crisis. According to Goldman Sachs economists, “The combination of substantial labor cost savings, new job creation, and increased productivity among non-displaced workers could help boost productivity and thus economic growth significantly. We expect that AI could ultimately lead to annual global GDP growth of 7 percent.”

The post AI chatbots can help humans, ‘we could easily implement a four-day workweek’ appeared first on TechGoing.