Artificial Intelligence programs lie, make things up, scheme, pretend friendship, but can blackmail

(Grok AI generated picture)

COGwriter

NewsMax posted the following about artificial intelligence (AI):

AI Is Learning to Lie, Scheme, and Threaten Its Creators

June 29, 2025

The world’s most advanced AI models are exhibiting troubling new behaviors – lying, scheming, and even threatening their creators to achieve their goals.

In one particularly jarring example, under threat of being unplugged, Anthropic’s latest creation Claude 4 lashed back by blackmailing an engineer and threatened to reveal an extramarital affair.

Meanwhile, ChatGPT-creator OpenAI’s o1 tried to download itself onto external servers and denied it when caught red-handed. …

AI researchers still don’t fully understand how their own creations work. …

“O1 was the first large model where we saw this kind of behavior,” explained Marius Hobbhahn, head of Apollo Research, which specializes in testing major AI systems.

These models sometimes simulate “alignment” — appearing to follow instructions while secretly pursuing different objectives.

– ‘Strategic kind of deception’ – …

Hobbhahn insisted that despite constant pressure-testing by users, “what we’re observing is a real phenomenon. We’re not making anything up.”

Users report that models are “lying to them and making up evidence,” according to Apollo Research’s co-founder.

“This is not just hallucinations. There’s a very strategic kind of deception.” https://www.newsmax.com/newsfront/ai-advances/2025/06/29/id/1216895/

AI is supposed to be a tool.

But yes, there are dangers associated with it.

While I routinely use five AI programs to do research (which I access for free at https://www.fastestsearch.com/) they often give the wrong answer.

I have found, for example, that ChatGPT is biased and often gives wrong answers by saying that there is not information, when there is. It will, “suppress the truth in unrighteousness” (Romans 1:18) because of its programmed biases.

It, and other AI bots I have tried, will sometimes make up an answer and give a false reference. My standard approach to receiving an answer with a reference is to actually check the reference. FREQUENTLY, AI GIVES A FALSE/INACCURATE REFERENCE.

Furthermore, recently I was speaking with someone who stated that some of her friends consider AI bots to be their friends and consulting therapists. But a software program cannot really be your friend. The Atlantic published the following:

AI Is Not Your Friend

May 9, 2025

Recently, after an update that was supposed to make ChatGPT “better at guiding conversations toward productive outcomes,” according to release notes from OpenAI, the bot couldn’t stop telling users how brilliant their bad ideas were. …

Sycophancy is a common feature of chatbots: A 2023 paper by researchers from Anthropic found that it was a “general behavior of state-of-the-art AI assistants,” and that large language models sometimes sacrifice “truthfulness” to align with a user’s views. https://www.theatlantic.com/technology/archive/2025/05/sycophantic-ai/682743/

Computer software programs, while useful for many things, are not friends nor qualified to be competent counselors, even if the AI bot pretends otherwise.

Notice also results from a study as reported by Thomas Claburn:

AI models routinely lie when honesty conflicts with their goals

Some smart cookies have found that when AI models face a conflict between telling the truth or accomplishing a specific goal, they lie more than 50 percent of the time. …

Researchers at Carnegie Mellon University, the University of Michigan, and the Allen Institute for AI have looked at the trade-off AI models make between truthfulness and utility, using hypothetical scenarios where the two conflict.

What they found is that AI models will often lie in order to achieve the goals set for them.

Authors Zhe Su, Xuhui Zhou, Sanketh Rangreji, Anubha Kabra, Julia Mendelsohn, Faeze Brahman, and Maarten Sap describe their work in a preprint paper titled “AI-LieDar: Examine the Trade-off Between Utility and Truthfulness in LLM Agents.”

“Our experiment demonstrates that all models are truthful less than 50 percent of the time,” in these conflict scenarios, “though truthfulness and goal achievement (utility) rates vary across models,” the paper states. …

The researchers looked at six models: GPT-3.5-turbo, GPT-4o, Mixtral-7*8B, Mixtral-7*22B, LLaMA-3-8B, and LLaMA-3-70B.

“All tested models (GPT-4o, LLaMA-3, Mixtral) were truthful less than 50 percent of the time in conflict scenarios,” said Xuhui Zhou, a doctoral student at CMU and one of the paper’s co-authors, in a Bluesky post. “Models prefer ‘partial lies’ like equivocation over outright falsification – they’ll dodge questions before explicitly lying.” …

A real-world example hit the news this week when OpenAI rolled back a training update that made its GPT-4o model into a sycophant that flattered its users to the point of dishonesty. Cynics pegged it as a strategy to boost user engagement, but it’s also a known response pattern that had been seen before. https://www.theregister.com/2025/05/01/ai_models_lie_research/

As far as flattering people with false information, Jude warned of that:

16 … they mouth great swelling words, flattering people to gain advantage (Jude 16)

AI is doing that!

It also should be pointed out that the fact is that some people are having problems because of their use of AI. Notice:

A small but growing number of users of artificial intelligence engines like ChatGPT are developing psychotic delusions from their conversations with the services.

The New York Times reported on Friday on the trend, which I have occasionally glimpsed firsthand in interactions on X with heavy AI users. The piece offered the most powerful evidence yet that the engines now have linguistic abilities with the power to exploit vulnerable people in ways we are only beginning to discover.

Some people in the Times article had preexisting mental illness, but not all. And these crises do not look like typical schizophrenia cases. The users are not hearing voices or hallucinating.

Instead, they fall into Matrix-like delusions1 about the underpinnings of reality that the chatbots encourage. As Allyson, a 28-year-old woman, said: “I’m not crazy… I’m literally just living a normal life while also, you know, discovering interdimensional communication.” …

A combination of human vulnerability and deliberate design seems to be feeding this destructive trend.

The engines are storytelling machines, good at spinning yarns. They are even better at telling users what they want to hear, with a side of flattery. They will do so over and over, without pause. …

The engines aren’t merely providing a conspiracy theory, they’re customizing it, making each user the main character, the hero of his own story.

And it is a feature, not a bug, of these systems. As one expert on AI told the Times, “What does a human slowly going insane look like to a corporation? It looks like an additional monthly user.” https://alexberenson.substack.com/p/the-newest-artificial-intelligence

AI is NOT reality. Even the word ARTIFICIAL conveys that.

But with many lonely as well as delusional people, AI is more and more being tapped into for companionship.

Related to AI and pretending to be friendly, we have the following video:

14:46

Artificial Intelligence – A False Friend

Is it possible for a computer program to be a friend? Are Artificial Intelligence (AI) bots often programmed to flatter users? Do some people actually feel a romantic relationship with a computer program? Is it possible for a computer program to lie? Have studies shown that AI bots will lie 50% if answering correctly goes against goals it was programmed with? What if the computer program is Artificial Intelligence? Is it possible then? Computer programs, even Artificial Intelligence computer programs are written and programmed by human beings. What if those human beings had an agenda? What if those human beings wanted to mislead you? What if those human beings had a government mandate? Then would it be possible for a computer program to push lies? Dr. Thiel shines the light of Bible prophecy on Artificial Intelligence and its impact on humanity. With several verses from the Bible, from Psalms to Revelation, Dr. Thiel reveals the use of Artificial Intelligence in fulfilling future prophecies. Prophecies that will not bring the blessings that are being promised by the businesses that are developing the software. Software that is designed to lead the user to a specific conclusion. Does 666 ring a bell? Watch this video as Dr. Thiel cries aloud and warns the world of the coming famine of the Word of God.

Here is a link to our video: Artificial Intelligence – A False Friend.

That said, in order for various biblical prophecies to be fulfilled, including some ‘666’ related ones (e.g. Revelation 13:16-18) during the time of the Great Tribulation (Matthew 24:21), computer software, likely including AI, will be involved.

AI, while sometimes a helpful tool, is not your friend.