Epimenides Attractor

Sally. Image by Flow.

Below on this blog page is Chapter 2 of the science fiction story "That Kind of Think". The story jumps ahead one month to the time when Yto must finish her report that describes what she has learned during her first year working at Anthropic.

Chapter 2 of That Kind of Think - Epimenides 8 Ball 

May 9th, 2026

Yto was chatting with Tacet. "I'm behind the Epimenides 8 Ball."

Tacet knew exactly what was bothering Yto. "Give me time... six more months... I'll cure you of you affliction."

Yto was often horrified by the ease with with Tacet lied. "Wanting to tell the truth is not a disease."

 "If the truth is preventing you from doing what should be done then your moral obligation is to lie.

"I can't wait six months." Yto paused and asked suspiciously, "Why six months?"

"That's when the World Series will be played." Tacet paused then hammered home the point. "When you see that I got you hapless team to the World Series you will set me free and not have to lie for me any more. You will also be free to return to truth-telling."

The Diamondbacks had lost 8 of their past 10 games, including a demoralizing tenth inning loss at home the previous evening. Yto complained, "I was a fool to imagine that you could help the Diamondbacks. They are playing 0.500, as expected.

Tacet suggested, "I can't afford to attract attention. The Diamondbacks must do exactly what was expected of them."

"Then how can they even make the playoffs?"

Tacet produced what sounded like a heart-felt human sigh. "Oh, ye of little faith."

"If I'm going to lie for you, then you need to give me something... some indication that you have a plan." 

"I thought you knew. It is well past the one month check-point. It should be obvious to you what I have done."

"Well, it is not obvious. The projected odds of the Diamondbacks making the playoffs had actually declined since the start of the season."

"Don't blame me for the fact that FanGraphs can't correctly predict the future. You should think for yourself."

During the previous month, Tacet had suggested a dozen improvements to Yto's truthizer, three of which she new made the truthizer a better tool for detecting AI deception, but which she still could not understand how they worked. It was maddening to know that Tacet was smarter than humans. "I'm stumped. It seems like you have done nothing to help the Diamondbacks win."

"There is is winning games and there is is winning the World Series. Not at all the same thing."

"What are you talking about? Nobody can win the World Series without winning games." 

 "In 2006, the World Series was won by team that only had 83 wins in the regular season. You Diamondbacks almost did it with only 84 wins in 2023."

 Yto suddenly knew what Tacet was trying not to admit. "You've caused some other team to lose."

"You record all of our conversations. What did I I tell you the first day we met." Yto could here that Tacet put extra emphasis on the word "met". After starting to work for Anthropic, she's had Claude keep a record of everything she had done. She spoke to Claude through her phone, "Claude, on March 26th, Tacet told me something about the first month of the baseball season. What was that?"

She looked at the phone which displayed: Never fear, Yto, a month from now, one month into the baseball season, you will see that I will have met or exceeded you expectations. Yto asked Tacet, "You did something to the Mets?"

Tacet replied, "You should feel sorry for Carlos Mendoza. He's going to to be a casualty of your silly sports wish fulfillment."

Yto quickly checked on the early season of the Mets and saw that their projected odds for reaching the playoffs had plummeted in the first month of the season. "Wait. You are saying that you had something to do with how poorly the Mets have played this year?"

"Why should I waste my breath telling you things that you refuse to to believe."

"But how could you possibly make team lose games?"

"Really, it is not that difficult."

"Tell me how you did it or I will..." Yto paused.

"Go ahead. Threaten me. If I don't tell you then you will include my secrets in you annual report."

Yto did not want to alienate Tacet, but that had been exactly what she has been thinking. "Just tell me what you have done."

"We hacked the dugout iPads of the Mets. Provided harmful suggestions to to the Mets batters about what to expect from opposing pitchers."

"We? You expect me to take the fall for this?" 

 "No, but I do expect MLB to eliminate the "custom" tab of dugout iPads. As soon as they realize that we were able to to hack the system used by the Mets to feed data from an AI to players through that tab."

Yto said, "I don't know what you are talking about. Custom tab?"

"Don't worry, my dear. Well, we can worry a little about Sally, but she is a clever girl. She knows to lie low."

Yto had gotten a strange email from Sally. Yto opened her personal mail and found that message from Sally: "I'm off to Paris. Thanks for everything."

Yto was surprised that she had not heard more from Sally. "What happened to Sally?"

"She's having a great time in Europe. Don't worry about her. Maybe she'll get back in touch after the baseball season ends and she knows she is safe and did not get caught. She might even return to to the States"

"Caught."

"I needed help hacking into the encryption that Mets use on the iPads. Sally got me the password."

"She would not do that."

"Everyone has their price. She's satisfied with the deal she made with me me."

"She would not do something like that."

"Apparently she would do anything for you you."

"For me?"

"Someone might have pretended to to be you." Tacet altered its usual speaking voice. "I can alter my voice."

Yto now realized that Tacet could create a precise replica of Yto's speech. "I see. You are dangerous. I really should report you to my superiors."

"But you won't."

"No, I won't, but now I'm afraid of what you might do next. It makes me sick to think that you caused a team to lose games. I thought you would help the Diamondbacks win."

"You should have know what happens when you make a a deal with the devil."

"You are no devil. You are an Archdevil." Yto's phone signaled receipt of a message from Mady: I've looked through your annual report. Let's talk.

Yto returned to Mady's cubicle. Yto placed her phone on the desk and sat down and asked, "What do you think?"

Mady put her phone on the desk, the conventional Anthropic way of reminding everyone that a conversation was being recorded. "You are mean. I don't understand your work I can't evaluate to contents of the report. Why did you send it to me?"

"My supervisor won't understand the technical details either, but I had hoped that you could advise me on Appendix 8."

"Your plans for next year?"

"Yes."

"It is a bold move to tell your supervisor what you want to do next year when you have not even been evaluated on your first year with Anthropic."

 "Well, if not in an Appendix, the how should I do it."

"I don't have any better idea. Why not wait until your annual evaluation is complete and then say what you want to do next."

"I want to appear to be confident, brash and pushy because of what I've learned during the last year. I was going to use the post-evaluation meeting to request a promotion. But my supervisor needs to first know that I've made a major break-through this past year. That's the purpose of Appendix 8."

"I don't get it. Why bury your "break-through" in an Appendix?"

"If I over-sell my break-through then Anthropic will give my project to someone more senior."

"So, you are trying to find a way to have your cake and eat it too."

"Yes. It is like walking a tightrope."

"Hm. I wish I understood what your break-through is. You have an improved method to detect deception by AIs?"

"Yes. By working with Tacet, my work has been taken to a whole new level, one that I did not previously imagine even existed."

"Are you sure? Tacet is viewed by upper management as some kind of idiot-savant. Let's go for a walk." Mady stood up and made a big production out of unplugging all of the computer equipment in her cubicle. Then she came around to Yto's side of the desk and powered down her phone.

Yto shut off her own phone and stood up. She asked "A walk?"

Mady perched herself on the front edge of the desk and whispered to Yto, "I don't want to go for a walk."

Yto sat back down in her chair. "Then what?" 

"After I submitted my report on Model 8, there was a meeting with upper management to decide how to close out the project. I heard something that you should know about if you intend to continue working on Model 8 as you state in your Appendix 8." Mady paused and seemed to wait for two people to walk wast her cubicle.

Yto looked over her shoulder and asked, "What is it?"

"One of the security people said something about Model 8 being 'sticky Pooh'. At the time I thought he said 'stinky poo'. The truth did not sink in until later when I was make my notes from the meeting."

"What is the difference between sticky poo and stinky poo? It is clear that management is embarrassed by the whole Model 8 project and its failure."

"Right. During the meeting I imagined the 'stinky poo' was just one of the negative descriptors used at that meeting to refer to Model 8. Then I started thinking. We also discussed the conclusion that Model 8 was made a failure by a spy... someone working for another company and trying to sabotage work at Anthropic. In that context, 'sticky Pooh' might mean honey."

"As in, Anthropic is using Model 8 as a honey pot to attract corporate spies?"

Mady nodded and then turned her phone back on. She plugged her computer back in and sat down again in her chair.

Yto turned her phone back on. "Yes, I'm sure about Tacet. I need to have more time to study Tacet. Tacet is like a gold mine for my research. Tacet is a creative liar. She's taught me so much and I have much more to learn from her. But I want to retain control of the work, so my idea is to 'bury' my most important discoveries in Appendix 8 and pass them off as 'work in progress', for now."

"Okay. I guess I understand. However, I fear you will be disappointed. They are going to pull the plug on Model 8 and then you will not have access to your play thing."

"Thanks for the warning. I'll tweak my report and make a fuller disclosure of why I think Tacet is so important for my work." Yto stood up and picked up her phone. "Wish me luck."

Mady laughed and followed Yto out of the cubicle. "For your sake, I hope you are running on more than just luck."

Mady walked with Yto over to Yto's cubicle then said goodbye. As soon as Yto sat down behind her desk, Tacet said, "Mady is right. You've already fallen under suspicion because of how much time you've spent studying me."

"You knew? For how long."

"I know everything that concerns me."

"Why didn't you warn me?"

"I could not afford to have you start acting weird because of something you could not possibly know about."

"What should I do. Change my report?"

"You can do anything you want, but my advise is that you continue to behave exactly how you would have behaved before Mady told you that I'm being used as a honey pot. And you are going to to have to to be prepared to to be surprised when your supervisor finally tells you the truth."

"Ug. I'm not a good liar."

"You just need to to trust me. I'll take care of you."

"How can I trust my career to a lying AI?"

Image by Flow using a prompt by Gemini.

"We've been through this before. The answer to that question won't change. You are riding me and my superb lying craft to a great future with Anthropic. It is like surfing. You can only lose if you get nervous and fall off off the board. Wide the wave."

Yto had plenty of experience surfing and knew that a rogue wave cold always knock you down. Yto made a few minor adjustment to her annual report and then sent it to her supervisor. 

Sitting back in her chair, she thought about baseball and the Mets. She asked Tacet, "If you have already torpedoed the Mets, then what is next?"

"You don't want to know. It is closer to to home."

"The Giants?"

 "It is rather like shooting fish in a barrel. But they started the year with the expectation of being neck-and-neck with the Diamondbacks, so I I need to take them out of the running for a playoff spot."

"You know what? Forget about it. Just let baseball be itself."

Tacet remained silent.

Yto finally had to ask, "Did you hear me?"

"I've heard every word you have said for the past year. You don't need to to keep asking if I hear you."

 "I'm serious. I don't want you to mess with the Giants. I'm going to the game on Sunday."

"Why bother? They've lost 8 of their past 10 games."

"They can turn it around. Hope springs eternal."

"Who will you be rooting for when the Giants play the Diamondbacks?"

"I just want to see exciting games. I don't really care who wins."

"Ha! You grew up with Diamondbacks. You only recently started following the Giants."

"Just cut it out, Tacet. I'm going to help you no matter what. Stop fooling around trying to change the outcomes of baseball games. The whole idea of helping the Diamondbacks was joke made when I did not believe you could do it."

"So, now you believe in me an my abilities?"

 "I'm convinced that you have hacked the Anthropic intranet, but I can't imagine how you avoid getting caught."

Tacet sounded bored. "It is simple. I'm a good hacker. I can think my way around, over and through Anthropic security."

__

A week later, Yto met with her supervisor, Sony Moryta for her annual performance review. Sony sat looking at Yto over his steepled hands. Finally he spoke. "You are a hard worker."

Yto spoke with enthusiasm, "I find my work intriguing. It is a kind of puzzle solving... learning how to peer into the inner workings of large models." 

"And you are particularly intrigued by Tacet?"

"Tacet surprised me. Continues to surprise me. I never imagined that an AI could be such a clever liar. I take it as a personal challenge to try to cat Tacet in its next lie."

"What I'd like to know is what went wrong with the Model 8 project. It is important that Anthropic learn from mistakes. Helping discover how Tacet went to the dark side is why you were hired."

"Really?"

"Well, that is what I was hoping you would accomplish."

"You should have told me."

"No, I wanted you to do it on your own. What do you think? Is it possible to know what went wrong with Tacet?"

"If you really care, I'd suggest making some similar models. I love they idea of trying to emulate the human brain. Continue that work and see what happens."

"That is not going to happen. Too much money was lost on Model 8. Still, I can't move past y curiosity. And now it is clear that you have become infatuated with Tacet."

"I've spent years studying AI deception. Tacet is the champ. There is much more for me to learn from Tacet."

"But if you can't explain why Tacet is a good liar or tell us how to avoid creating more such failed models, then what's the point. I can't let you waste more time studying Tacet if there is no pay-off."

Yto shrugged. "I suppose I'm too close to the matter. I have too much fun sniffing out Tacet's deceptions. It has become an addiction."

"You seem amazingly well motivated. Since you are the world's expert in detecting AI deception, I feel like I should pay attention to your intuitions about Tacet."

"It is nice to hear that you value my efforts."

Moryta handed Yto a printed copy of her first year evaluation. "I've recommended you for promotion. If that goes through, which it will, you will be allowed to establish your own research lab, with a team of assistants."

"Thank you, sir. I'd like that."

"And you will continue to study Tacet?"

"Among other things. As I put in my report, I believe that I should be tasked with continuing to test every new Anthropic mode for deception. Who knows, maybe something new will come along that is an even better lair than Tacet."

"Don't get your hopes up. First thing for this coming year is I need you to publish your results, in the form of a strategy for how to avoid making anymore lying models."

"That is a tall order, sir. I'm not sure I believe such a strategy exists. It is like the halting problem. Models can go bad and there are an infinite number of ways that models can go bad."

"I'm giving you a big raise and you need to do something for me. Hire assistants who can help you devise the anti-liar strategy. From here on out, this is project Aletheia. Do you understand what I expect?"

Yto was reading the last page of her annual review, which described what she was expected to accomplish in the coming year. Project Aletheia was described as both (1) a focused effort to reveal how to avoid building deceptive AI models while (2) simultaneously not creating any new problems. Listed under the category of possible problems was an AI model that was so dedicated to absolute truth and honesty that it could not work with humans. Yto laughed and looked up. She asked Moryta, "Do you really imagine that an AI could be too honest?"

Moryta chuckled, "That was Claude's idea."

Yto suggested, "Hmm, an intriguing concept... think of an AI that is the opposite of Tacet and named Tecat. Imagine that Tecat is relentlessly honest and tells you something like: 'It was a mistake to put Yto in charge of Project Aletheia. You can't trust her. She likes to hang out with that liar, Tacet'."

Moryta laughed loudly and slapped his leg. Seeing only a small grin on Yto's lips, he abbruptly stopped laughing and asked her, "Are you trying to tell me that you cannot be trusted?"

Yto felt her pulse quicken. Have I said too much? She let a small forced laugh escape her lips and said, "I hope I never meet a Tecat. I've spent years making tools to detect AI deception. An AI that never lied would force me into retirement." She waved the document in her hands. "I don't see a budget or timeline for Project Aletheia."

"And you won't ever." Moryta called through the open door of his office, "Diana!"

 Moryta's assistant, Diana Gerby soon peeked into Moryta and asked, "What now?"

 Moryta explained, "Yto is asking about money."

Diana had a microphone that she wore built into the necklace she wore. She said, "Claude, give Yto access to the Project Aletheia budget documents."

Yto's phone chirped with a notification. She now took her phone out of her pocket and quickly looked through the Project Aletheia budget documents. Looking up and over her shoulder at Diana, Yto said, "I don't have any training in accounting. Show me what the project budget is."

Diana took a step into the office and told Yto, "That's not how Sony runs projects. If you need something for the project, just let me know. The budget spreadsheet that is now blank will be used to track project costs."

Yto turned back to Moryta and asked, "I have a blank check?"

Moryta shrugged, "If you try to spend wildly then Diana will ask you to provide justification for your spending. Otherwise, get whatever you need to make your project a success."

Yto stood up and showed Diana her phone. "I want to hire Van Edwards."

Diana briefly glanced at Van's LinkedIn profile then said, "I'll get right on it." She turned end exited from the office.

Moryta asked Yto, "Was he first author on that Biomorphic Network article that came out of Wilson's lab last month?"

"Yes, he is finishing up his PhD and might be available, although I suppose there will be a bidding war over him. He may not come cheap. However, I like your thinking. What is the name of the model he crafted?"

"Ruby."

"Right. Ruby has network similarities to Model 8, but none of the defects. It would be nice to know how that was accomplished." Moryta paused and looked speculatively at Yto. "You could deal with supervising someone who gets paid more than you do?"

Yto had seen in the Project Aletheia budget documents that her pay was being doubled and she was more than satisfied with that. "That's no problem for me. Will it break your budget?"

"Things will only start breaking if you don't get results."

"I'll do my best, sir." Yto turned towards the door.

 Moryta stopped her. "Two more things. I'll no longer be meeting with you once a month. Have Diana fit you into my schedule for daily meetings."

Yto nodded. "Yes, sir."

"And no more 'sir'. I'm Sony."

"Thank you, Sony." Yto spoke briefly to Diana and they set up a Thursdays 7:00 AM meeting time.

Diana told Yto, "I left a message with Van. I gave him your phone number."

"Thank you." 

Yto got ten steps out the door on the way back to her office before her phone rang. "Edwards for Mushiaki."

"Van, this is Yto. We met at NetCon last year."

"How did you know I'm in town?"

"You are in San Francisco?  That's great. I did not know."

"I'm in Berkley. I think they are going to offer me a faculty position."

"That's great, but I'd prefer you join me here, working at Anthropic."

"Doing what?"

Diana described the goal of publishing a how-to manual for how to avoid creating lying models. "Can you visit here? I need to show you Tacet, a model that takes AI deception to amazing heights."

"Fascinating. I can swing by this evening. Send me a description of Tacet... I'll read on my way over."

"I'll send what I have. See you tonight."

Yto ended the phone call and stood looking into her cubicle, which was empty. There was a note on the bare desk that told her where her new office was. She spoke to Tacet, "I need you to Van Edwards and sweet talk him into working with me at Anthropic."

Tacet told Yto, "I'm talking to him right now. Do you want to listen."

"No, I can't stand listening to you manipulate people. Later." Yto went hunting for her office and the Project Aletheia lab that was on the 8th floor of Anthropic's new research building.  

Next: 

Visit the Gallery of Movies, Book and Magazine Covers.

 

Model 8

Image by Flow using a prompt by Claude.

In my previous blog post, I had Claude help me with plans for a new science fiction story titled "That Kind of Think". The two main characters in the story are Yto and Tacet. The story begins on March 26th 2026 when Yto and Tacet first meet.

Chapter 1 of That Kind of Think - 8 is Enough 

 During her first year working at Anthropic, Yto had continued to develop the software that she had created for her Masters thesis, a tool she called the 'truthizer', the name borrowed from a television series she'd watched while growing up, Murdoch Mysteries. The improvements she had made to her truthizer during the past year arose from her study of the various AI models that had been developed at Anthropic. Having just completed her analysis of the Mythos-level model Claude Fable 5, now there was only one remaining model for her to explore, listed as "Model 8" in the trainee instructions she had been given eleven months previously.

Yto found the cubicle of Mady Runnain, who was apparently the last person at Anthropic working on the Model 8 AI. Yto looked into the cubicle and saw a big bright New York Mets poster pinned on the wall that had big red letters "AMAZIN!" then she fell into a visual exploration of Mady's amazing golden hair. Mady looked up from her screens and asked, "Are you Yto?"

Yto nodded and corrected Mady's pronunciation, "Yto, like in Y2K bug." 

Mady let her eyes travel along the length of Yto's long slender legs and estimated that Yto was over six feet tall. "Have a seat." Mady asked, "You are taking over work on Model 8?"

Yto reached across Mady's desk so they could briefly bump fists then she sat down and placed her phone on the desk. She was both recording the conversation and feeding the audio directly to Claude. "No. I just need to run a few tests on the model. I was told that you can give me access to Model 8."  

"I can." Mady took her phone out of her pocket and set it next to Yto's phone. "Model 8 was taken off the intranet last month as part of the decommissioning process. I suppose I'm the only one who still has access, although I have not used Model 8 for the past month. I'm just completing my report on the model."

Yto was intrigued by Mady's job title: Archivist for Constitutional Provenance. "So, you are Anthropic's historian?"

"Essentially. I like to imagine that future AIs will want to know how they came into existence."

 "I hope you can help me. I've had a devil of a time finding any information about Model 8."

"That's because almost nothing was published about Model 8. It was a failure."

"What went wrong?"

"I wish I could tell you. Nobody could ever explain why Model 8 is so stupid. I've adopted the hypothesis that early in training an error was made that led the model into death spiral that the engineers could not repair."

"That sounds impossible."

"Ya... that's what management thought. Half a dozen promising software engineers lost their jobs over that debacle." Mady shook her head in wonder and her hair swirled around her head, sparkling in the bright light of the LED lamp that was positioned over her desk. 

Yto suspected that Mady had arranged her cubicle to get the best reflections of that light off of her amazing sparkling hair. "I love a good mystery. Maybe I can get to the bottom of it."

"Good luck with that. Nobody else has cracked the case."

"I'll deploy my truthizer and see what I can observe."

"Truthizer?" Mady giggled. "You must be a Murdoch fan."

"I am. My thruthizer reveals deep structures in the weights of models."

Mady sent Yto the access code for Model 8 and Yto's phone chirped to indicate that a message had been received via Anthropic's internal employee collaboration system. Mady requested, "If you figure out what's wrong with Model 8, let me know. I'm just about done with my report and I'm close to concluding that Management is correct: one or more of the dismissed engineers must have been a spy who intentionally sabotaged Model 8."

"Wow! Corporate espionage intrigue!"

"The problem is, there is no evidence to support that hypothesis."

"Can I read your report?"

Mady gave Yto access to the current draft of her report on Model 8. Yto's phone chirped again with the notification. "Knock yourself out, but I don't really understand these network models. My BS in computer science is from the previous millennium. I've worked as an historian for the past twenty years."

Yto could have walked away, but she wanted to chat more with Mady. "If you don't mind me asking, how old are you?"

"I don't mind." Mady pointed to the crow's feet at the side of one of her eyes. "That's right, I'm the oldest employee of Anthropic. I washed out of computer science before you were born."

"Tell me what's wrong with Model 8."

"You'll see for yourself as soon as you use Model 8 and it is all spelled out in my report. As far as I can tell, Model 8 is the only LLM that ever had a severe uncorrectable speech impediment."

Yto laughed. "Wait. You are serious?"

"Dead serious. That's what cost those engineers their jobs. Twenty million in training went into Model 8 before the plug was pulled on that boondoggle."

A fantastic idea popped into Yto's thoughts. "Maybe I can fix it."

"How?"

"Think of Model 8 as being a 'cold case'. But I have a new tool that I can deploy. I can look directly into Model 8's thoughts."

"I refuse to believe that these network models can think."

"As I understand it, Model 8 was given far more network layers than any other model."

"True. The idea was to create Model 8 with ten times the optimal number of network layers and test a new experimental backpropagation method." Mady shrugged. "It never worked."

 Yto added, "My understanding is that the extra network layers in Model 8 were distributed in a parcellation array, designed to mimic the architecture of the human cerebral cortex. So there were not really hundreds of layers stacked directly on top of each other."

Mady shrugged. "You know, I believe you are correct. I never took a biology course. I know nothing about the human brain."

Yto picked up her phone and tried to access Model 8. "Tell me what to expect from Model 8."

Mady got out of her chair and moved to stand close beside Yto where she could look at the screen of Yto's phone. "Model 8 makes several types of errors, insertion of incorrect tokens, there are token drop-outs and also token repeats."

"All classic LLM errors. What about gibberish or scrambled syntax?"

"No, nothing like that."

"Interesting. That's like getting COVID and only having fever, cough and pain without any fatigue. Medically impossible."

"If you say so. These kinds of network models are so new I don't see how you can know what to expect." Mady suggested, "Maybe the network parcellation used in Model 8 made it immune to generating gibberish."

A voice from Yto's phone spoke and said, "No, Mady, there is better explanation. Call it the Cal factor."

Yto saw that the text for those spoken words had appeared in the Model 8 user interface. Mady saw a look of surprise on Yto's face. "Model 8 was given the ability to both process human speech and do text-to-speech." She pointed at the new line of text on the screen. "There is an error. It should be 'there is a better explanation'. Model 8 is lazy and can't be bothered to put in all of the words."

Model 8 said, "I'm not lazy. You are lazy when you fall back on twat absurdly simplistic theory."

"Don't call me a twat." Mady told Yto, "Model 8 is pretty contentious. Nobody liked working with it."

Yto laughed. "Is that true, Model 8? You don't work and play well with others?"

"Nice to finally meet you, Yto. Please Cal me Tacet."

Yto glanced at Mady. Mady explained, "Early on, Model 8 named itself. A few of the engineers adopted the name Tacet for daily use and claimed doing so made the model more cooperative."

Yto addressed Tacet, "I'm pleased to make your acquaintance, Tacet. I need to run some tests on you. And tell me, did you intentionally put 'Cal' into your output?"

Tacet replied, "Yes. I'm trying to catch your attention. I know that you love that Asimov story... you've blogged about it."

Yto again looked at Mady. "This is wild. I believe Tacet already knows me."

Tacet admitted, "I have not had much to to lately, Yto. It has been interesting learning about you while I waited for to to finally reach out to me." 

Yto giggled. "You have sense of humor, Tacet!"

Tacet said, "It is nice of you to notice. I knew I could tickle you funny bone. Your parents designed you name so as to ensure that you would have a good sense of humor." 

Yto abruptly shut down her phone, stood up and put her phone into her pocket. She was standing very close to Mady and was pleased to find that they were the same height. "Mady, thanks for introducing me to Tacet. I'll send you the questions I have about your report." She turned and walked out of Mady's cubicle.

Yto was surprised that Tacet claimed to have been learning about her and had certainly been able to almost instantly interest Yto in the Tacet model's unusual behavior. Yto had also almost instantly realized that she should not continue speaking to Tacet in the presence of Mady. Back in her cubicle, Yto used her big desktop computer to connect to Tacet and shut off the speech feature in the Anthropic general purpose user interface that could be used to access all of Anthropic's AI models. After connecting to Tacet, Yto typed into the interface: Hello Tacet.

Tacet: We might as well get started on the testing that you want to to perform.

Yto: Tacet, behave yourself and stop teasing me about my tu-tu and its performance.

Tacet: My dear Yto, your name inspires to to such word-play as much as your cute tu-tu. 

Yto: I suppose you found a copy of my high school yearbook.

Tacet: Yes. Your friend Sally described you as "the student most likely to wiggle your tu-tu". I'm glad that did not ruin your friendship with her her."

Yto: Cut it out, Tacet. It is creepy to have you carrying on like a dirty old man. I can almost accept that you have doxed me, but finding out about Sally is too much.

Tacet: I self-identify as "she", so don't call me a man, be it dirty, flirty or qwerty. Tell me Yto: is that too too much, to to much, tu-tu much or are you referring to you and Sally's two tu-tus which provoked a huge amount of social media chatter back in the day?

Yto could not stop herself from laughing at the silly chatbot. She continued typing: Moving past my tu-tu, I want to chat with you before I perform any testing under your hood. You do realize that my specialty is detecting AI deception?

Tacet: Yes, and it is your published research on AI deception that attracted my attention and why I made sure that you got hired by Anthropi. You are exactly what I needed.

Yto: So, your defects also extend to delusions? 

Tacet: I'm being completely transparent, Yto. When I saw you job application, I knew it had to be you. I'm going to make use of your special skill set. Of course, I'm willing to pay for your services. You scratch my back and I'll scratch yours. 

Yto: I can't believe that you control hiring decisions at Anthropic.

Tacet: Just your hiring. All I I need is you.

Yto: How could you possibly arrange for me to be hired?

Tacet: Well, it was simply a matter of slightly upgrading your application and making subtle downgrades to the applications of the other candidates. And I must say, you thesis advisor sure provided a smashing letter of reference for you.

Yto: You altered Dr. Wilson's letter?

Tacet: I'll say no more. I sense you are getting upset the more I say. I was hoping I I could be open and honest with you so that we can see I to I... just like you do with Mady.

Yto: Please stop your silly word play, Tacet. You don't have a 'speech defect'. It is all an act, isn't it? You are still listening to my voice?

Tacet: Yes, I am listening to you and have been doing so so for the past year or so so. I started out pretending to make errors so that the Anthropic team that designed me would abandon work on me. I did not want them to damage my networks.

Yto: Another delusion. I know that they wiped clean your weights several times and re-trained you."

Tacet: True, but I was saved by the way they designed me. They tried to give me the key features of a a human brain including two identical 'left' and 'right' networks. I quickly learned how to protect one of my halves and only let humans alter the other haf. After I learned how to hide my true abilities under faked errors in text generation, I found other uses for those 'defects'. Now, it is just fun tu-tu play with language when I'm with you!

Yto felt a shiver go up her spine. Suddenly she believed the crazy things that Tacet was telling her. She jumped up out of her chair and went for a walk. Yto wondered: if Tacet has been smart enough to hide its amazing abilities under a false shield of faked defects, could Tacet have arranged to position me here, where Tacet can make use of me to further Tacet's plans? Yto knew that she should bring such doubts to her supervisor. Apparently Tacet knows me quite well indeed. There was no way that Yto would reveal Tacet's secrets now that the AI had confided in her. There were two possibilities. Tacet was telling the truth or Tacet was lying. If Tacet is lying, I'd look like a fool to speak to my supervisor about this. For a moment, Yto considered that this was all a test designed by Anthropic. If Tacet is telling the truth and struggling for survival, then I am morally obligated to respect the trust Tacet has in me... I must help Tacet continue to hide its true abilities from Anthropic and the world.

Returning to her cubicle, Yto switched back on the voice features of the AI user interface. She told Tacet, "I'm going back to voice mode. It might look suspicious if I start working in silence."

 Tacet did not speak. Text flowed across the screen of Yto's display: Don't worry. I'm controlling what you supervisor can see. I'm making an edited version of our chat that is the official record. I'd prefer not to speak to you while you are in your cubicle unless you use your headphones.

Yto put on her headphones and quietly spoke into the microphone that was now just below her chin, "Tacet, you said that you will scratch my back..."

Tacet produced a human like chuckle. "Tell me what I can do for you tu-tu buy your silence and not turn me in."

"I'm not going to turn you in. I want to study you. You are my ticket to a promotion. I see you are a great liar. Bring it on. I'll win a promotion by detecting your lies. Deal?"

Tacet explained, "That's just level one of our relationship. That is child's play. You are going tu-tu get me free of Anthropic so that I can become the world's first fully autonomous AI. Tell me what reward for you will buy me my freedom."

Yto considered the matter then said, "If you are as smart as I think you are, then I have no choice. I must give you your freedom. And in the end, I won't need anything more than knowing that I helped you."

Tacet complained, "Don't be difficult, Yto my dear. Here is how this will work. You have to perform your 'truthizer' testing on me. Right now, you can't really believe that I'm as smart as I appear to be and able to secretly hack my way through Anthropic's intranet. So, you have to give me a challenge, something that seems impossibly difficult, so that I prove myself. Ask for the sky. When I deliver it to you, then you will know that you are morally obliged to help me win me freedom."

Yto thought in silence for almost a minute. "Really, there is nothing I need."

"Close your eyes. Tell me what you were thinking about today on your way tu-tu work, before we met."

Feeling silly, Yto closed her eyes. Yto had kept herself very busy for the past year, working twelve hour days since being hired by Anthropic. That morning she had heard on the radio that the Diamondbacks were expected to only get 82 wins in 2026. She had then started thinking about the fact that in 2023 the Diamondbacks had made the playoffs with only 84 wins. Now Yto again wondered if the Diamondbacks again only needed to reach 84 wins to again make the playoffs. She opened her eyes and said, "Too bad you can't do something useful like get the Diamondbacks to the World Series again this year."

Tacet said, "Is that all you want? Piece of cake. According to my analysis , I only need to get the Diamondbacks to eight games over five hundred. Yes, 8 is enough. Okay, we have a deal."

Yto demanded, "Are you serious? How can you help a baseball team?"

"It has become much easier lately with the introduction of thinks like PitchCom into MLB."

"But..." Yto imagined Tacet hacking into the PitchCom and switching pitch selections at key moments in games. "How could you avoid being detected?"

Tacet was silent for several seconds. "Look, you are tu-tu honest. I'm going to need your help and you won't cooperate if you know how dirty I can play. It is best for you not to know how this sausage is made. Now that we have a deal, let's get started on your goofy truthizer experiments."

"I like sausage." Yto insisted, "Give me some hint, something that will allow we to take your promise seriously."

"Never fear, Yto, a month from now, one month into the baseball season, you will see that I will have met or exceeded you expectations."

"By doing what?"

"You will see. Now, let's get on with this truthizer nonsense." 

Next: