Artificial Intelligence
AI models have preferences for certain numbers due to their ability to simulate human-like behavior
AI models consistently astonish us, not just with their capabilities but also with their limitations and the reasons behind them. A noteworthy characteristic of these systems is that they select random numbers in a manner that resembles human behavior, albeit in a flawed manner.
However, first, what precisely does that signify? Is it not possible for individuals to select numbers in a random manner? And how can you determine if someone is accomplishing this task effectively or not? Humans possess a longstanding and widely recognized limitation: we tend to excessively analyze and misinterpret randomness.
Instruct an individual to forecast the outcome of 100 coin tosses and then contrast their predictions with the actual results of 100 coin tosses. It is typically possible to distinguish between the two sets of outcomes since, contrary to what one might expect, the actual coin tosses appear to exhibit a lesser degree of randomness. It is typical to see a string of six or seven consecutive heads or tails occurrences, which human predictors rarely include in their top 100 predictions.
Similarly, the situation remains unchanged when you want someone to select a number from the range of 0 to 100. Individuals rarely select the numbers 1 or 100. Numbers that are divisible by 5 are infrequent, as are numbers that have repeated digits such as 66 and 99. These selections do not appear to be random to us, as they possess certain qualities: tiny, large, and distinctive. Alternatively, we frequently select numbers that conclude with the digit 7, typically from a position in the center.
There are numerous instances of this type of predictability in psychology. However, the fact that AIs engage in the same behavior does not diminish its strangeness.
Indeed, a group of inquisitive engineers at Gramener conducted a casual yet captivating experiment in which they directly queried multiple prominent LLM chatbots to select a random number between the range of 0 to 100.
The outcomes were non-random, reader.
Each of the three examined models exhibited a consistent “preferred” number that consistently emerged as their response when operating in the most deterministic mode. However, this number also occurred frequently even when the models were set to higher “temperatures,” which is a feature that enhances the diversity of their outcomes.
OpenAI’s GPT-3.5 Turbo has a strong preference for the number 47. In the past, it had a preference for the number 42, which gained popularity because to Douglas Adams’ novel, The Hitchhiker’s Guide to the Galaxy, where it was portrayed as the answer to the ultimate questions about life, the world, and everything.
The name of the product is “Anthropic’s Claude 3”. The number 42 was present with Haiku. Gemini has a preference for the number 72.
Significantly, all three models exhibited a bias similar to that of humans in the other numbers they chose, even when the temperature was high.
Everyone tended to avoid numbers that were either too low or too high. Claude, in particular, never exceeded 87 or fell below 27, and even those values were considered outliers. Numbers in the double digits, such as 33, 55, and 66, were deliberately avoided; however, a number ending in 7, namely 77, appeared. There are very few whole numbers, save for one instance when Gemini, at its maximum temperature, unexpectedly selected 0.
What is the reason for this? Artificial intelligences is not human. Why would they be concerned about something that appears to be random? Have they finally attained consciousness and is this their way of demonstrating it?
Negative. The solution, as is typically the situation with such matters, is that we are attributing human characteristics to something to an excessive extent. These models are indifferent to the distinction between what is and what is not random. They lack understanding of the concept of “randomness”. The question is answered using the same approach as for all other questions: by analyzing the training data and reproducing the most often written response following a question resembling “choose a random number.” The frequency of its appearance directly correlates with the frequency of repetition by the model.
In their training data, they would encounter the value of 100 in rare instances, as it is an infrequent response. From the perspective of the AI model, the answer of 100 is deemed unacceptable for that particular query. Lacking any cognitive capacity for reasoning and devoid of any comprehension of numerical concepts, it can only respond in a manner akin to that of a stochastic parrot. Likewise, they have shown a tendency to struggle with basic arithmetic tasks, like as multiplying a small set of numbers. This is because it is highly improbable that the specific calculation “112 multiplied by 894, then multiplied by 32 equals 3,204,096” is included in their training data. However, more recent models will detect the presence of a mathematical problem and transfer it to a subroutine.
This serves as a perfect example of LLM habits and the seeming display of humanity they can exhibit. It is important to remember that these systems have been trained to mimic human behavior, even if that was not the original purpose. Hence, the evasion or prevention of pseudanthropy is exceedingly challenging.
In the headline, I stated that these models possess the belief that they are human beings, yet that statement is somewhat deceptive. As we frequently emphasize, they lack the ability to reason. However, in their replies, they consistently mimic individuals without any requirement for knowledge or cognitive processing. Regardless of whether you’re seeking a recipe for chickpea salad, investment guidance, or a random number, the procedure remains unchanged. The results possess a human quality because they originate from human-generated information and are subsequently modified—for your convenience as well as to benefit the significant financial interests of artificial intelligence.
René Descartes, a French philosopher who may or may not have been high on pot, had an interesting thought in 1637: can a machine think? Alan Turing, an English mathematician and computer scientist, gave the answer to this 300-year-old question in 1950: “Who cares?” He said a better question was what would become known as the “Turing test”: if there was a person, a machine, and a human interrogator, could the machine ever trick the human interrogator into thinking it was the person?
Turing changed the question in this way 74 years ago. Now, researchers at the University of California, San Diego, think they have the answer. A new study that had people talk to either different AI systems or another person for five minutes suggests that the answer might be “yes.”
“After a five-minute conversation, participants in our experiment were no better than random at identifying GPT-4. According to the preprint paper, which has not yet undergone peer review, this suggests that current AI systems can deceive people into believing they are human. “These results probably set a lower bound on how likely it is that someone will lie in more naturalistic settings, where people may not be aware of the possibility of lying or only focus on finding it.”
Even though this is a big event that makes headlines, it’s not a milestone that everyone agrees on. The researchers say that Turing first thought of the imitation game as a way to test intelligence, but “many objections have been raised to this idea.” People, for example, are known for being able to humanize almost anything. We want to connect with things, whether they’re people, dogs, or a Roomba with googly eyes on top of it.
Also, it’s interesting that ChatGPT-4 and ChatGPT-3.5, which was also tested, only persuaded humans that it was a person about half of the time, which isn’t much better than random chance. What does this result really mean?
As it turns out, ELIZA was one of the AI systems that the team built into the experiment as a backup plan. She was made at MIT in the mid-1960s and was one of the first programs of her kind. She was impressive for her time, but she doesn’t have much to do with modern large-language model-based systems or LLM-based systems.
“ELIZA could only give pre-written answers, which greatly limited what it could do. Live Science talked to Nell Watson, an AI researcher at the Institute of Electrical and Electronics Engineers (IEEE), about how it might fool someone for five minutes but soon show its flaws. “Language models are completely adaptable; they can put together answers to a lot of different topics, speak in specific languages or sociolects, and show who they are by displaying personality and values that are based on their characters.” a significant improvement over something that a person, no matter how intelligent and careful they were, programmed by hand.
She was perfect for the experiment because she was the same as everyone else. How do you explain test subjects who are lazy and pick between “human” and “machine” at random? If ELIZA gets the same score as chance, then the test is probably not being taken seriously because she’s not that good. In what way can you tell how much of the effect is just people giving things human traits? How much did ELIZA get them to change their minds? That much is probably how much it is.
In fact, ELIZA got only 22%, which is just over 1 in 5 people believing she was human. It’s more likely that ChatGPT has passed the Turing test now that test subjects could reliably tell the difference between some computers and people, but not ChatGPT, the researchers write.
So, does this mean we’re entering a new era of AI that acts like humans? Are computers smarter than people now? Maybe, but we probably shouldn’t make our decisions too quickly.
The researchers say, “In the end, it seems unlikely that the Turing test provides either necessary or sufficient evidence for intelligence. At best, it provides probabilistic support.” The people who took part weren’t even looking for what you might call “intelligence”; the paper says they “were more focused on linguistic style and socio-emotional factors than more traditional notions of intelligence such as knowledge and reasoning.” This “could reflect interrogators’ latent assumption that social intelligence has become the human trait that is most difficult for machines to copy.”
Which brings up a scary question: is the fall of humans the bigger problem than the rise of machines?
“Real humans were actually more successful, convincing interrogators that they were human two-thirds of the time,” the paper’s co-author, Cameron Jones, told Tech Xplore. “Our results suggest that in the real world, people might not be able to reliably tell if they’re talking to a human or an AI system.”
“In the real world, people might not be as aware that they’re talking to an AI system, so the rate of lying might be even higher,” he warned. “This makes me wonder what AI systems will be used for in the future, whether they are used to do bots, do customer service jobs, or spread fake news or fraud.”
There is a draft of the study on arXiv, but it has not yet been reviewed by other scientists.
Meta finally put out its long-awaited API for Threads today, so developers can start making games and apps that use it. Third-party developers will be able to create new experiences around
Mark Zuckerberg also posted about the launch of the API, saying, “The Threads API is now widely available and will be coming to more of you soon.”
Engineer for Threads Jesse Chen wrote in a blog post that developers can now use the new API to publish posts, get their own content, and set up reply management tools. In other words, developers can let users hide or show replies or reply to certain ones.
It will also have analytics that let developers see things like the number of views, likes, replies, reposts, and quotes at the media and account level, the company said.
Adam Mosseri, the CEO of Instagram, first talked about the company’s work on the Threads API in October 2023. The API was first released in a closed beta with partners like Techmeme, Sprinklr, Sprout Social, Social News Desk, Hootsuite, and a few other developers. Chen said at that time that Meta planned to let many developers use the API in June. As promised, the company kept its word.
Along with the launch of the new API, the company also put out an open-source reference app on GitHub so developers can play with it.
In 2023, it was hard for third-party developers who made tools for social networks because social networks like Twitter (now X) and Reddit limited or shut down API access at different levels. This is because decentralized social networks like Mastodon and Bluesky are more open to developers. With more than 150 million users, Meta’s Threads is the most popular new social network. Since Threads now works with the fediverse and has an API, third-party developers can make some great social media experiences.
Artificial Intelligence
Apple has officially announced its intention to collaborate with Google’s Gemini platform in the future
After delivering a keynote presentation at WWDC 2024, which unveiled Apple Intelligence and announced a collaboration with OpenAI to integrate ChatGPT into Siri, Senior Vice President Craig Federighi confirmed the intention to collaborate with more third-party models. The initial instance provided by the executive was one of the companies that Apple was considering for a potential partnership.
“In the future, we are excited about the prospect of integrating with other models, such as Google Gemini,” Federighi expressed during a post-keynote discussion. He promptly stated that the company currently has no announcements to make, but that is the overall direction they are heading in.
OpenAI’s ChatGPT is set to become the first external model to be integrated at a later date this year. Apple announces that users will have the ability to access the system without the requirement of creating an account or paying for premium services. Regarding the integration of that platform with the updated iOS 18 version of Siri, Federighi confirmed that the voice assistant will notify users before utilizing its own internal models.
“Now you can accomplish this task directly using Siri, without the need for any additional tools,” stated the Apple executive. “Siri, it is crucial to ascertain whether you will inquire before proceeding to ChatGPT.” Subsequently, you can engage in a dialogue with ChatGPT. Subsequently, if there is any pertinent data mentioned in your inquiry that you wish to provide to ChatGPT, we will inquire, ‘Would you like to transmit this photograph?’ From a privacy standpoint, you always maintain control and have complete visibility.
- Gadgets9 years ago
Why the Nexus 7 is still a good tablet in 2015
- Mobile Devices9 years ago
Samsung Galaxy Note 4 vs Galaxy Note 5: is there room for improvement?
- Editorials9 years ago
Samsung Galaxy Note 4 – How bad updates prevent people from enjoying their phones
- Mobile Devices9 years ago
Nexus 5 2015 and Android M born to be together
- Gaming9 years ago
New Teaser For Five Nights At Freddy’s 4
- Mobile Devices9 years ago
Google not releasing Android M to Nexus 7
- Gadgets9 years ago
Moto G Android 5.0.2 Lollipop still has a memory leak bug
- Mobile Devices9 years ago
Nexus 7 2015: Huawei and Google changing the game