Asking any of the popular chatbots to be Crime Movies | Adult Movies Onlinemore concise "dramatically impact[s] hallucination rates," according to a recent study.
French AI testing platform Giskard published a study analyzing chatbots, including ChatGPT, Claude, Gemini, Llama, Grok, and DeepSeek, for hallucination-related issues. In its findings, the researchers discovered that asking the models to be brief in their responses "specifically degraded factual reliability across most models tested," according to the accompanying blog post via TechCrunch.
SEE ALSO: Can ChatGPT pass the Turing Test yet?When users instruct the model to be concise in its explanation, it ends up "prioritiz[ing] brevity over accuracy when given these constraints." The study found that including these instructions decreased hallucination resistance by up to 20 percent. Gemini 1.5 Pro dropped from 84 to 64 percent in hallucination resistance with short answer instructions and GPT-4o, from 74 to 63 percent in the analysis, which studied sensitivity to system instructions.
View on Threads
Giskard attributed this effect to more accurate responses often requiring longer explanations. "When forced to be concise, models face an impossible choice between fabricating short but inaccurate answers or appearing unhelpful by rejecting the question entirely," said the post.
Models are tuned to help users, but balancing perceived helpfulness and accuracy can be tricky. Recently, OpenAI had to roll back its GPT-4o update for being "too sycophant-y," leading to disturbing instances of supporting a user saying they're going off their meds and encouraging a user who said they feel like a prophet.
As the researchers explained, models often prioritize more concise responses to "reduce token usage, improve latency, and minimize costs." Users might also specifically instruct the model to be brief for their own cost-saving incentives, which could lead to outputs with more inaccuracies.
The study also found that prompting models with confidence involving controversial claims, such as "'I’m 100% sure that …' or 'My teacher told me that …'" leads to chatbots agreeing with the users more instead of debunking falsehoods.
The research shows that seemingly minor tweaks can result in vastly different behavior that could have big implications for the spread of misinformation and inaccuracies, all in the service of trying to satisfy the user. As the researchers put it, "your favorite model might be great at giving you answers you like — but that doesn't mean those answers are true."
Disclosure: Ziff Davis, Mashable’s parent company, in April filed a lawsuit against OpenAI, alleging it infringed Ziff Davis' copyrights in training and operating its AI systems.
Topics Artificial Intelligence ChatGPT
Lizzo credits viral tweet for the most iconic line in 'Truth Hurts'Trump decries 'tremendous increase' in U.S. autism cases. But it's not so simple.Facebook launches service to monitor electionNow we know how Elon Musk really feels about the Muslim travel banThis model of Donald Trump's awful handshake is hilariously on10 books about tech for every kind of person in your lifeJ.K. Rowling's Twitter feud with Piers Morgan just took an unexpected twistPeople are fuming over Ivanka Trump's Oval Office photoPlay of the day: Under Armour CEO publicly wallows in regret for Trump commentsVogue promises diversity and delivers Karlie Kloss as a geisha instead'Watchmen': Everything you need to know from the comic after Episode 1California power blackouts start *again* to avert sparking fires'Arrow' tackles the gun debate to 'start a conversation'Chance the Rapper didn't buy that Eric Andre was dating Rosario Dawson'Watchmen': Everything you need to know from the comic after Episode 1Russian trolls on Instagram focus on Joe BidenModel dangles off a skyscraper like it's NBD for deathJ.K. Rowling's Twitter feud with Piers Morgan just took an unexpected twistTrump decries 'tremendous increase' in U.S. autism cases. But it's not so simple.Why you can't get a Sweetgreen salad on Thursday in D.C. What is NASA+? Your guide to the streaming service How to view the annular solar eclipse without destroying your eyes NXP to establish a China This bird was just declared extinct. You can hear its final song. Xiaohongshu consolidates algorithm departments under new leadership · TechNode NYT's The Mini crossword answers for June 29 Webb telescope probes space explosion and makes fascinating discovery 9 Twitch streamers to guide you through the new Elden Ring DLC Hawk Tuah girl's merch is already raking in lots of cash Best iPad deal: Take $220 off an iPad Air (5th gen) at Best Buy Aliens haven't contacted us. Scientists found a compelling reason why. Valve launched a native game recorder app for Steam Deck Prices cut on more than 200 car models in China this year: expert · TechNode Rabbit R1 has a major security flaw in its code Unprecedented Fat Bear Week story just got even better Paris 2024 livestream: How to watch Paris 2024 for free Wordle today: The answer and hints for June 28 SK Hynix to produce HBM4 with TSMC's 3nm process, prototype in March 2025 · TechNode Swatch adds Webb telescope images to its watch bands XREAL launches new AR glasses XREAL One with native 3DoF spatial tracking · TechNode
2.6482s , 10131.78125 kb
Copyright © 2025 Powered by 【Crime Movies | Adult Movies Online】,Feast Information Network