ChatGPT's Voice Mode has some security flaws,redhead teen sex videos but OpenAI says it's on top of it.
On Thursday OpenAI published a report on GPT-4o's safety features, addressing known issues that occur when using the model. GPT-4o is the underlying model that powers the latest version of ChatGPT, and comes with a Voice Mode that was recently released to a select group of users with a ChatGPT Plus subscription.
SEE ALSO: What OpenAI's Scarlett Johansson drama tells us about the future of AIThe "safety challenges" identified include standard risks like prompting the model with erotic and violent responses, other disallowed content, and "ungrounded inference" and "sensitive trait attribution" — assumptions that might be discriminatory or biased, in other words. OpenAI says it has trained the model to block any outputs flagged in these categories. However, the report also says mitigations don't include "nonverbal vocalizations or other sound effect" such as erotic moans, violent screams, and gunshots. One can infer, then, that prompts involving certain sensitive nonverbal sounds might improperly receive a response.
OpenAI also mentioned unique challenges that come with vocally communicating with the model. Red-teamers discovered that GPT-4o could be prompted to impersonate someone or accidentally emulate the user's voice. To combat this, OpenAI only allows pre-authorized voices (minus the notorious Scarlett Johansson-sounding voice). GPT-4o can also identify other voices besides the speaker's voice, which presents a serious privacy and surveillance issue. But it has been trained to deny those requests — unless the model is being prompted on a famous quote.
Red-teamers also noted that GPT-4o could be prompted to speak persuasively or emphatically, a feature that could be more harmful than text outputs when it comes to misinformation and conspiracy theories.
Notably, OpenAI also addressed potential copyright issues that have plagued the company and the overall development of generative AI, which trains on data scraped from the web. GPT-4o has been trained to refuse requests for copyrighted content and has additional filters for blocking outputs containing music. On that note, ChatGPT's Voice Mode has been directed not to sing under any circumstances.
OpenAI's numerous risk mitigations covered in the lengthy document were carried out before Voice Mode was released. So the ostensive message of the report says that while GPT-4o is capable of certain risky behavior, it won't do it.
However, OpenAI says, "These evaluations measure only the clinical knowledge of these models, and do not measure their utility in real-world workflows." So it's been tested in a controlled environment, but when the broader public gets their hands on GPT-4o, it could be a different beast when out in the wild.
Mashable reached out to OpenAI for additional clarity about these mitigations, and will update if we hear back.
Topics Artificial Intelligence OpenAI
Ebina Named Winner of 'America's Got Talent'
Get schooled online by celebrities like Steve Martin, Shonda Rhimes
Chinese woman gets her wish to undergo gender confirmation surgery at 72
Racist posters targeting Chinese students stir controversy at the University of Texas at Austin
LTHS Launches Short Story Contest
Forget graphing calculator hacks, real cheaters use a tortoise
Vladimir Putin HATES this meme so much Russia banned it
LGBTQ community responds perfectly after being accused of stealing God's rainbow
Strumming Along
Rocket launch marks first key test for NASA moon base
2013 VJCC Summer Festival to Take Place June 22
A different giraffe gave birth, in case you need your giraffe birth fix
Housemates reach new levels of petty with a note for their Amazon deliveryman
Uber driver picks up her boyfriend's other girlfriend and then sh*t really hits the fan
Free Screening of ‘The Crumbles’ at JANM
Trump's election data firm doesn't think his White House looks that fun
Here's how women are represented (or not) in LinkedIn's highest
Forget graphing calculator hacks, real cheaters use a tortoise
Crossing Cultures
The White House says it's totally saving Trump's deleted tweets, don't you worry
Steel, Kim Win Second Term in CongressA Path of Panther Greatness: Lauren Kamiyama Inducted into Chapman University Sports Hall of Fame'Squid Game' Gets 3 SAG Awards$50,000 Reward Announced for Unsolved Murder in Woodland HillsNikkei Basketball Heritage Assn. Awarded Grant to Interview Former Camp InmatesResults for Asian American Candidates in Assembly, Court of Appeal, County RacesShort Men 'Don't Have Human Rights', says eSports Pro on StreamOBITUARY: Bruce Asakawa, Major Figure in Gardening WorldJANM Statement on Passage of Amache National Historic Site ActJapan PM Orders Probe of Unification Church Problems LGBTQ Flag Burned at Pasadena Buddhist Temple; Police Investigating Incident as Hate Crime West Covina Cherry Blossom Festival Next Saturday Japanese Lunar Lander Apparently Crashes Amache National Historic Site Invites Public Input on Development of Foundation Document? Strengthening Memory Through Rhythm New Netflix Death Note Series to Be Led by Creators of Stranger Things Biden Proclaims 2023 AANHPI Heritage Month Anime Expo Announces Anime Expo Ontario ‘Free Chol Soo Lee’ Screening at Hammer Museum Google Doodle Celebrates Photographer Corky Lee
0.1581s , 8443.828125 kb
Copyright © 2025 Powered by 【redhead teen sex videos】ChatGPT Voice Mode is capable of some freaky stuff — but here's how OpenAI is tackling it.,Global Hot Topic Analysis
