Many Large Language Model (LLM) chatbots are designed and used for companionship, and people have reported forming friendships, mentorships, and romantic partnerships with them. Concerns that companion chatbots may harm or replace real human relationships have been raised, but whether and how these social consequences occur remains unclear. In the present longitudinal study ($N = 183$), participants were randomly assigned to a chatbot condition (text chat with a companion chatbot) or to a control condition (text-based word games) for 10 minutes a day for 21 days. Participants also completed four surveys during the 21 days and engaged in audio recorded interviews on day 1 and 21. Overall, social health and relationships were not significantly impacted by companion chatbot interactions across 21 days of use. However, a detailed analysis showed a different story. People who had a higher desire to socially connect also tended to anthropomorphize the chatbot more, attributing humanlike properties to it; and those who anthropomorphized the chatbot more also reported that talking to the chatbot had a greater impact on their social interactions and relationships with family and friends. Via a mediation analysis, our results suggest a key mechanism at work: the impact of human-AI interaction on human-human social outcomes is mediated by the extent to which people anthropomorphize the AI agent, which is in turn motivated by a desire to socially connect. In a world where the desire to socially connect is on the rise, this finding may be cause for concern.

Word games hold significant research value for natural language processing (NLP), game theory, and related fields due to their rule-based and situational nature. This study explores how large language models (LLMs) can be effectively involved in word games and proposes a training-free framework. "Shei Shi Wo Di" or "Who is the Spy" in English, is a classic word game. Using this game as an example, we introduce a Chain-of-Thought (CoT)-based scheduling framework to enable LLMs to achieve excellent performance in tasks such as inferring role words and disguising their identities. We evaluate the framework's performance based on game success rates and the accuracy of the LLM agents' analytical results. Experimental results affirm the framework's effectiveness, demonstrating notable improvements in LLM performance across multiple datasets. This work highlights the potential of LLMs in mastering situational reasoning and social interactions within structured game environments. Our code is publicly available at https://github.com/ct-wei/Who-is-The-Spy.

My task for our AI overlords was simple: help me crack the New York Times Spelling Bee. I had spent a large chunk of a Saturday evening trying to shape the letters G, Y, A, L, P, O and N into as many words as possible. But three hours, 141 points and 37 words -- including "nonapology", "lagoon" and "analogy" -- later, I had hit a wall. A few more words was all I needed to propel myself into Spelling Bee's "genius" echelon, the title reserved for those who unscramble 70 percent of all possible words using the given letters, and the point at which the puzzle considers itself, effectively, solved. My human mind was clearly struggling, but this task seemed like child's play for AI, so I fired up ChatGPT, told it I was trying to win the Spelling Bee, gave it my letters and laid out the rules.

The recent breakthrough in large language models (LLMs) such as ChatGPT has revolutionized every industry at an unprecedented pace. Alongside this progress also comes mounting concerns about LLMs' susceptibility to jailbreaking attacks, which leads to the generation of harmful or unsafe content. While safety alignment measures have been implemented in LLMs to mitigate existing jailbreak attempts and force them to become increasingly complicated, it is still far from perfect. In this paper, we analyze the common pattern of the current safety alignment and show that it is possible to exploit such patterns for jailbreaking attacks by simultaneous obfuscation in queries and responses. Specifically, we propose WordGame attack, which replaces malicious words with word games to break down the adversarial intent of a query and encourage benign content regarding the games to precede the anticipated harmful content in the response, creating a context that is hardly covered by any corpus used for safety alignment. Extensive experiments demonstrate that WordGame attack can break the guardrails of the current leading proprietary and open-source LLMs, including the latest Claude 3, GPT 4, and Llama 3 models more effectively than existing attacks efficiently. Further ablation studies on such simultaneous obfuscation in query and response provide evidence of the merits of the attack strategy beyond an individual attack. Warning: The paper contains unfiltered text generated by LLMs which can be offensive.

It's been a terrible year for game developers, but an amazing year for games. There were some missteps along the way -- if you'd asked me to predict this list a year ago, I would've mentioned both Redfall and Starfield -- but overall it's been a packed year unusually low on disappointment. We've never tried to name a single title as "the Game of the Year." Instead, it's become a tradition to get the whole team together to talk about our individual favorites. So here are those games, presented in alphabetical order to avoid hurting any of our writers' feelings. Feel free to sound off about what your favorites are in the comments; there are no wrong answers. I rarely have time to finish games these days, but I devoured Alan Wake 2 in just a few weeks. For me and my limited gaming time, that felt miraculous. I'll admit, I'm a mark for Remedy Entertainment. I've been following its work since the first Max Payne arrived on PCs in 2001, right as I was gearing up to head to college and building my first desktop PC. Yah, I was one of the cool kids on campus..) Max Payne blew me away with its fluid slow-motion gunplay mechanics and immersive narrative. As a lifelong console gamer until then, it was a big step forward from something like Tomb Raider.

Crossword puzzles are one of the most popular word games, played in different languages all across the world, where riddle style can vary significantly from one country to another. Automated crossword resolution is challenging, and typical solvers rely on large databases of previously solved crosswords. In this work, we extend WebCrow 2.0, an automatic crossword solver, to French, making it the first program for crossword solving in the French language. To cope with the lack of a large repository of clue-answer crossword data, WebCrow 2.0 exploits multiple modules, called experts, that retrieve candidate answers from heterogeneous resources, such as the web, knowledge graphs, and linguistic rules. We compared WebCrow's performance against humans in two different challenges. Despite the limited amount of past crosswords, French WebCrow was competitive, actually outperforming humans in terms of speed and accuracy, thus proving its capabilities to generalize to new languages.

When you move across the Atlantic armed with only one year of self-taught Portuguese and a joie de vivre that's been chewed up and spit out by two years of a pandemic, things in your day-to-day life can start to feel a little … desperate. One day you're laughing at your own confusion and the next you're, I don't know, sobbing comically in a café bathroom because you ordered a latte and got bread instead. Let's face it--regardless of whether or not you're feeling a little lost in Europe's westernmost country, big wins are few and far between for any of us these days. In comes Wordle, a simple game with predictable rules, daily wins, a mediated dopamine rush, and a lighter reason to check in with your loved ones. And god was it a hit.

Player 1 first starts with a word in mind that is at least 4 letters long, e.g. He plays the first letter of the word, i.e. 'm'.