The intel came from a leaked audio file of an internal presentation on an early version of Microsoft’s Security Copilot a ChatGPT-like artificial intelligence platform that Microsoft created to assist cybersecurity professionals.
Apparently, the audio consists of a Microsoft researcher addressing the result of "threat hunter" testing, in which the AI examined a Windows security log for any indications of potentially malicious behaviour.
"We had to cherry-pick a little bit to get an example that looked good because it would stray and because it's a stochastic model, it would give us different answers when we asked it the same questions," said Lloyd Greenwald, a Microsoft Security Partner giving the presentation, as quoted by BI.
"It wasn't that easy to get good answers," he added.
Security Copilot, like any chatbot, allows users to enter their query into a chat window and receive responses as a customer service reply. Security Copilot is largely built on OpenAI's GPT-4 large language model (LLM), which also runs Microsoft's other generative AI forays like the Bing Search assistant. Greenwald claims that these demonstrations were "initial explorations" of the possibilities of GPT-4 and that Microsoft was given early access to the technology.
Similar to Bing AI in its early days, which responded so ludicrous that it had to be "lobotomized," the researchers claimed that Security Copilot often "hallucinated" wrong answers in its early versions, an issue that appeared to be inherent to the technology. "Hallucination is a big problem with LLMs and there's a lot we do at Microsoft to try to eliminate hallucinations and part of that is grounding it with real data," Greenwald said in the audio, "but this is just taking the model without grounding it with any data."
The LLM Microsoft used to build Security Pilot, GPT-4, however it was not trained on cybersecurity-specific data. Rather, it was utilized directly out of the box, depending just on its massive generic dataset, which is standard.
Discussing other queries in regards to security, Greenwald revealed that, "this is just what we demoed to the government."
However, it is unclear whether Microsoft used these “cherry-picked” examples in its to the government and other potential customers – or if its researchers were really upfront about the selection process of the examples.
A spokeswoman for Microsoft told BI that "the technology discussed at the meeting was exploratory work that predated Security Copilot and was tested on simulations created from public data sets for the model evaluations," stating that "no customer data was used."
Ambitious Warcraft fans have persuaded an AI article bot into writing about the mythical character Glorbo in an amusing and ingenious turn of events. The incident, which happened on Reddit, demonstrates the creativity of the game industry as well as the limitations of artificial intelligence in terms of fact-checking and information verification.
The hoax gained popularity after a group of Reddit users decided to fabricate a thorough backstory for a fictional character in the World of Warcraft realm to test the capabilities of an AI-powered article generator. A complex background was given to the made-up gnome warlock Glorbo, along with a made-up storyline and special magic skills.
The Glorbo enthusiasts were eager to see if the AI article bot would fall for the scam and create an article based on the complex story they had created. To give the story a sense of realism, they meticulously edited the narrative to reflect the tone and terminology commonly used in gaming media.
To their delight, the experiment was effective, as the piece produced by the AI not only chronicled Glorbo's alleged in-game exploits but also included references to the Reddit post, portraying the character as though it were a real member of the Warcraft universe. The whimsical invention may be presented as news because the AI couldn't tell the difference between factual and fictional content.
The information about this practical joke swiftly traveled throughout the gaming and social media platforms, amusing and intriguing people about the potential applications of AI-generated material in the field of journalism. While there is no doubt that AI technology has transformed the way material is produced and distributed, it also raises questions about the necessity for human oversight to ensure the accuracy of information.
As a result of the experiment, it becomes evident that AI article bots, while efficient in producing large volumes of content, lack the discernment and critical thinking capabilities that humans possess. Dr. Emily Simmons, an AI ethics researcher, commented on the incident, saying, "This is a fascinating example of how AI can be fooled when faced with deceptive inputs. It underscores the importance of incorporating human fact-checking and oversight in AI-generated content to maintain journalistic integrity."
The amusing incident serves as a reminder that artificial intelligence technology is still in its infancy and that, as it develops, tackling problems with misinformation and deception must be a top focus. While AI may surely help with content creation, it cannot take the place of human context, understanding, and judgment.
Glorbo's developers are thrilled with the result and hope that this humorous occurrence will encourage discussions on responsible AI use and the dangers of relying solely on automated systems for journalism and content creation.