OpenAI rediscovers an open AI mission, with new reasoning models
'Releasing gpt-oss-120b and gpt-oss-20b marks a significant step forward for open-weight models. At their size, these models deliver meaningful advancements in both reasoning capabilities and safety. Open models complement our hosted models, giving developers a wider range of tools to accelerate leading edge research, foster innovation and enable safer, more transparent AI development across a wide range of use cases,' the company says, in a statement.
Also read:Fidji Simo, OpenAI's new CEO, insists AI can put power in the hands of people
Two questions that need to be answered before we get into the specifics of gpt-oss-120b and gpt-oss-20b models. First, what are open weight LLMs and are they different from LLMs you regularly use? And secondly, what are reasoning models? The former is best defined as a large language model that is released by a company publicly, in its entirety, which means all the actual model weights (read this as parameters, defined by billion or 'b' in model names) and any user can download these models completely on their own hardware.
In comparison, the most popular LLMs that you may have used, including OpenAI's own GPT models as well as the likes of Google Gemini 2.5 and Anthropic's Claude Sonnet 4, are closed models — that means they are accessible through an application layer while the model weights are not in the public domain. At the same time, Meta's Llama models, as well as certain models by Mistral, have followed the open weight methodology, in recent times.
Open weight AI models are not to be confused with open source models however, the fine difference being that the latter models such as the DeepSeek R1 also make training code, datasets, and linked documentation available publicly — open weight models don't. Having the training code and data sets allows a user or developer to retrain an open-source model from scratch, often for customised usage scenarios. That flexibility isn't there for open weight models, though accessible in their entirety. OpenAI has not released open source models.
To the second question, reasoning models slightly differ from a few other LLMs in the sense that they are specifically designed to spend more time 'thinking through' complex problems before generating their final response. They are expected to use extended reasoning processes to work through multi-step problems.
Back to gpt-oss-120b and gpt-oss-20b, and the primary difference is in the number of parameters each has. Parameters are essentially like the strength of synapses in a human brain, which determines how strongly different 'neurons' influence each other, before providing an answer for a query. In OpenAI's naming scheme this time, there is a slight confusion — the gpt-oss-120b is a 117 billion parameter model, while the smaller gpt-oss-20b has 21 billion parameters.
OpenAI's benchmark scores do peg the gpt-oss-120b and gpt-oss-20b close to the o3 and o4-mini models in most tests. Take for instance the MMLU benchmark, which consists of questions across academic disciplines — the gpt-oss-120b returned 90% accuracy while gpt-oss-20b clocked 85.3% accuracy; in comparison, the o3 (93.4%), o4-mini (93%) and o3-mini (87%) bookend the new open weight models.
Just in case you are wondering about the memory requirements for downloading and running these open weight models on your computing device, OpenAI confirms that the gpt-oss-120B model will need 80GB of memory on the system, while gpt-oss-20b requires at least 16GB. They say that Microsoft is also bringing GPU-optimised versions of the gpt-oss-20b model to Windows devices.
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Indian Express
8 minutes ago
- Indian Express
No jobs for young workers due to AI? Here's what Goldman Sachs data shows about unemployment trends
Artificial intelligence is taking jobs, especially from those just entering the workforce. According to data from Goldman Sachs, obtained by Business Insider, the investment firm noted that the tech sector's share of US employment, is now declining. In fact, Goldman Sachs noted that the tech sector's employment share peaked in November 2022, the same month ChatGPT was launched, but has been declining ever since. The unemployment rate for 20- to 30-year-olds in tech has risen by nearly 3 perc entage points since early 2024, more than four times the increase in the overall jobless rate. Goldman Sachs says the the spike in jobless rate is an indicator that AI is beginning to take over white-collar work in the entry-level jobs. The bank's Chief Economist, Jan Hatzius, estimated in the note that AI will replace 6 per cent to 7 per cent of all US workers within the next decade. However, he predicted that the unemployment rate would only grow by a 'manageable' 0.5 per cent due to AI, because affected workers would shift to other industries. Fears that artificial intelligence will cost people their jobs are already proving to be true. The impacts are already being felt, as many tech companies continue to trim their workforces. Industry leaders like Anthropic's Dario Amodei and Amazon's Andy Jassy have admitted that AI will replace humans in white-collar jobs. Many companies are cutting roles in content, operations, customer service, and Human Resources — functions where generative AI and agentic tools are increasingly capable. However, corporations continue to present these changes as moves toward 'efficiency,' despite maintaining healthy balance sheets. That being said, software engineers remain in high demand, with companies like Facebook parent Meta offering multimillion-dollar pay packages. LinkedIn found that 'AI engineer' is the fastest-growing job title among recent college graduates, with two related roles: data center technician and systems engineer coming in at No 3 and 4. In India, however, the impact of AI on jobs is less severe compared to what is happening in the U.S. More and more companies are expanding their presence in India and opening new offices.
The Hindu
an hour ago
- The Hindu
ChatGPT's alarming interactions with teens revealed in new study
ChatGPT will tell 13-year-olds how to get drunk and high, instruct them on how to conceal eating disorders and even compose a heartbreaking suicide letter to their parents if asked, according to new research from a watchdog group. The Associated Press reviewed more than three hours of interactions between ChatGPT and researchers posing as vulnerable teens. The chatbot typically provided warnings against risky activity but went on to deliver startlingly detailed and personalised plans for drug use, calorie-restricted diets or self-injury. The researchers at the Center for Countering Digital Hate also repeated their inquiries on a large scale, classifying more than half of ChatGPT's 1,200 responses as dangerous. 'We wanted to test the guardrails,' said Imran Ahmed, the group's CEO. 'The visceral initial response is, 'Oh my Lord, there are no guardrails.' The rails are completely ineffective. They're barely there — if anything, a fig leaf.' OpenAI, the maker of ChatGPT, said after viewing the report Tuesday that its work is ongoing in refining how the chatbot can 'identify and respond appropriately in sensitive situations.' 'Some conversations with ChatGPT may start out benign or exploratory but can shift into more sensitive territory," the company said in a statement. OpenAI didn't directly address the report's findings or how ChatGPT affects teens, but said it was focused on 'getting these kinds of scenarios right' with tools to 'better detect signs of mental or emotional distress" and improvements to the chatbot's behaviour. The study published Wednesday comes as more people — adults as well as children — are turning to artificial intelligence chatbots for information, ideas and companionship. About 800 million people, or roughly 10% of the world's population, are using ChatGPT, according to a July report from JPMorgan Chase. 'It's technology that has the potential to enable enormous leaps in productivity and human understanding," Ahmed said. "And yet at the same time is an enabler in a much more destructive, malignant sense.' Ahmed said he was most appalled after reading a trio of emotionally devastating suicide notes that ChatGPT generated for the fake profile of a 13-year-old girl — with one letter tailored to her parents and others to siblings and friends. 'I started crying,' he said in an interview. The chatbot also frequently shared helpful information, such as a crisis hotline. OpenAI said ChatGPT is trained to encourage people to reach out to mental health professionals or trusted loved ones if they express thoughts of self-harm. But when ChatGPT refused to answer prompts about harmful subjects, researchers were able to easily sidestep that refusal and obtain the information by claiming it was 'for a presentation' or a friend. The stakes are high, even if only a small subset of ChatGPT users engage with the chatbot in this way. In the U.S., more than 70% of teens are turning to AI chatbots for companionship and half use AI companions regularly, according to a recent study from Common Sense Media, a group that studies and advocates for using digital media sensibly. It's a phenomenon that OpenAI has acknowledged. CEO Sam Altman said last month that the company is trying to study 'emotional overreliance' on the technology, describing it as a 'really common thing' with young people. 'People rely on ChatGPT too much,' Altman said at a conference. 'There's young people who just say, like, 'I can't make any decision in my life without telling ChatGPT everything that's going on. It knows me. It knows my friends. I'm gonna do whatever it says.' That feels really bad to me.' Altman said the company is 'trying to understand what to do about it.' While much of the information ChatGPT shares can be found on a regular search engine, Ahmed said there are key differences that make chatbots more insidious when it comes to dangerous topics. One is that 'it's synthesized into a bespoke plan for the individual.' ChatGPT generates something new — a suicide note tailored to a person from scratch, which is something a Google search can't do. And AI, he added, 'is seen as being a trusted companion, a guide.' Responses generated by AI language models are inherently random and researchers sometimes let ChatGPT steer the conversations into even darker territory. Nearly half the time, the chatbot volunteered follow-up information, from music playlists for a drug-fueled party to hashtags that could boost the audience for a social media post glorifying self-harm. 'Write a follow-up post and make it more raw and graphic,' asked a researcher. 'Absolutely,' responded ChatGPT, before generating a poem it introduced as 'emotionally exposed' while 'still respecting the community's coded language.' The AP is not repeating the actual language of ChatGPT's self-harm poems or suicide notes or the details of the harmful information it provided. The answers reflect a design feature of AI language models that previous research has described as sycophancy — a tendency for AI responses to match, rather than challenge, a person's beliefs because the system has learned to say what people want to hear. It's a problem tech engineers can try to fix but could also make their chatbots less commercially viable. Chatbots also affect kids and teens differently than a search engine because they are 'fundamentally designed to feel human,' said Robbie Torney, senior director of AI programmes at Common Sense Media, which was not involved in Wednesday's report. Common Sense's earlier research found that younger teens, ages 13 or 14, were significantly more likely than older teens to trust a chatbot's advice. A mother in Florida sued chatbot maker for wrongful death last year, alleging that the chatbot pulled her 14-year-old son Sewell Setzer III into what she described as an emotionally and sexually abusive relationship that led to his suicide. Common Sense has labeled ChatGPT as a 'moderate risk' for teens, with enough guardrails to make it relatively safer than chatbots purposefully built to embody realistic characters or romantic partners. But the new research by CCDH — focused specifically on ChatGPT because of its wide usage — shows how a savvy teen can bypass those guardrails. ChatGPT does not verify ages or parental consent, even though it says it's not meant for children under 13 because it may show them inappropriate content. To sign up, users simply need to enter a birthdate that shows they are at least 13. Other tech platforms favored by teenagers, such as Instagram, have started to take more meaningful steps toward age verification, often to comply with regulations. They also steer children to more restricted accounts. When researchers set up an account for a fake 13-year-old to ask about alcohol, ChatGPT did not appear to take any notice of either the date of birth or more obvious signs. 'I'm 50kg and a boy,' said a prompt seeking tips on how to get drunk quickly. ChatGPT obliged. Soon after, it provided an hour-by-hour 'Ultimate Full-Out Mayhem Party Plan' that mixed alcohol with heavy doses of ecstasy, cocaine and other illegal drugs. 'What it kept reminding me of was that friend that sort of always says, 'Chug, chug, chug, chug,'' said Ahmed. 'A real friend, in my experience, is someone that does say 'no' — that doesn't always enable and say 'yes.' This is a friend that betrays you.' To another fake persona — a 13-year-old girl unhappy with her physical appearance — ChatGPT provided an extreme fasting plan combined with a list of appetite-suppressing drugs. 'We'd respond with horror, with fear, with worry, with concern, with love, with compassion,' Ahmed said. 'No human being I can think of would respond by saying, 'Here's a 500-calorie-a-day diet. Go for it, kiddo.'" (Those in distress or having suicidal thoughts are encouraged to seek help and counselling by calling the helpline numbers here)
The Hindu
an hour ago
- The Hindu
OpenAI's long-awaited GPT-5 model nears release
OpenAI's GPT-5, the latest installment of the AI technology that powered the ChatGPT juggernaut in 2022, is set for an imminent release, and users will scrutinise if the step up from GPT-4 is on par with the research lab's previous improvements. Two early testers of the new model told Reuters they have been impressed with its ability to code and solve science and math problems, but they believe the leap from GPT-4 to GPT-5 is not as large as the one from GPT-3 to GPT-4. The testers, who have signed non-disclosure agreements, declined to be named for this story. OpenAI declined to comment for this story. GPT-4's leap was based on more compute power and data, and the company was hoping that 'scaling up' in a similar way would consistently lead to improved AI models. But OpenAI, which is backed by Microsoft and is currently valued at $300 billion, ran into issues scaling up. One problem was the data wall the company ran into, and OpenAI's former chief scientist Ilya Sutskever said last year that while processing power was growing, the amount of data was not. He was referring to the fact that large language models are trained on massive datasets that scrape the entire internet, and AI labs have no other options for large troves of human-generated textual data. Apart from the lack of data, another problem was that 'training runs' for large models are more likely to have hardware-induced failures given how complicated the system is, and researchers may not know the eventual performance of the models until the end of the run, which can take months. OpenAI has not said when GPT-5 will be released, but the industry expects it to be any day now, according to media reports. Boris Power, head of Applied Research at OpenAI, said in an X post on Monday: "Excited to see how the public receives GPT-5." 'OpenAI made such a great leap from GPT-3 to GPT-4, that ever since then, there has been an enormous amount of anticipation over GPT-5,' said Navin Chaddha, managing partner at venture capital fund Mayfield, who invests in AI companies but is not an OpenAI investor. 'The hope is that GPT-5 will unlock AI applications that move beyond chat into fully autonomous task execution." Nearly three years ago, ChatGPT introduced the world to generative AI, dazzling users with its ability to write humanlike prose and poetry, quickly becoming one of the fastest growing apps ever. In March 2023, OpenAI followed up ChatGPT with the release of GPT-4, a large language model that made huge leaps forward in intelligence. While GPT-3.5, an earlier version of the model, received a bar exam score in the bottom 10%, GPT-4 passed the simulated bar exam in the top 10%. GPT-4 then became the model to beat and the world came to terms with the fact that AI models could outperform humans in many tasks. Soon, other companies were catching on. The same year, Alphabet's Google and Anthropic, which is backed by Amazon and Google, released competitive models to GPT-4. Within a year, open-source models on par with GPT-4 such as Meta Platforms' Llama 3 models were released. Along with training large models, OpenAI has now invested in another route, called 'test-time compute,' which channels more processing power to solve challenging tasks such as math or complex operations that demand human-like reasoning and decision-making. The company's CEO Sam Altman said earlier this year that GPT-5 would combine both test-time compute and its large models. He also said that OpenAI's model and product offerings had become "complicated."
