GPT Protocol (GPT) news

Mentis captures signal changes in the US-Israel-Iran situation, AI Agent trading enters the "information-first" stage

MoltBook Exploded 4,700% in Just 24 Hours. Could Husky Inu AI Be Next?

Data Suggests Husky Inu AI Could Become One of the Few New Tokens Capable of a 20×–60× Surge Within Weeks of Listing

3 edging images, surged to the top of the Kaito Chinese community leaderboard in 24 hours

In general, for the InfoFi model to succeed, it will ultimately require a better balance between incentivizing deep content creation and maintaining its own self-sustaining capability.

OpenAI rolls back ChatGPT update after user concerns ignored

OpenAI Founder: Additional Fixes on Model's Personality Underway

OpenAI founder: GPT-4o has been updated to improve intelligence and personality

White House AI Plan Draws Over 10,000 Comments Amid Global Tech Race

The White House announced on April 24 that it had received more than 10,000 public submissions in response to its proposed artificial intelligence action plan, underscoring mounting national interest in the technology as global competition intensifies.

Anticipating ChatGPT 5’s Release: Potential Impact on Bitcoin Valuation

Anticipating ChatGPT 5: Its potential influence on Bitcoin and the role of Sam Altman's updates in shaping cryptocurrency fortunes

OpenAI’s new GPT-4.1 gives more unsafe and biased responses

Share link:In this post: GPT-4.1 is showing more unsafe and biased behavior than its predecessor, GPT-4o, in independent tests. OpenAI skipped its usual safety report for GPT-4.1, prompting researchers to investigate its reliability. Security tests reveal GPT-4.1 is easier to misuse due to its need for very clear instructions and poor handling of vague prompts.

OpenAI’s new ChatGPT models found to “hallucinate” more often

Share link:In this post: OpenAI’s new o3 and o4‑mini models hallucinate more than older versions, breaking past improvement trends. Tests show o3 and o4‑mini make up facts up to twice as often, with o4‑mini hallucinating nearly half the time. Real-time search may help reduce errors, but the root cause of rising hallucinations remains unknown.

Auradine Secures $153M in Series C Funding to Expand Blockchain and AI Infrastructure

Auradine, a prominent blockchain and AI infrastructure solutions provider, has successfully raised $153 million in a Series C funding round.

OpenAI releases 1 million token coding model GPT 4.1, available immediately via API

OpenAI founder: GPT-4.1 (and -Mini and -Nano) now available in the API

Stanford HAI "2025 AI Index Report" Highlights

Artificial Intelligence is becoming more efficient, accessible, and affordable; additionally, Asians are more optimistic about Artificial Intelligence.

OpenAI officially announced that GPT-4 will be retired at the end of this month and will be completely replaced by 4o

Meta’s LLaMA 4 Strengthens U.S. Lead In The Global AI Race

Meta's Llama 4 boosts US position in AI competition

Meta debuts Llama 4 AI model family with multimodal capabilities

Share link:In this post: Meta released Llama 4 Scout and Maverick on April 5 to boost its lead in generative AI. Llama 4 models can handle text, images, audio, and video and are freely available outside the EU. Meta plans to spend $65 billion on AI infrastructure and has expanded Llama 4 to 40 countries.

Sentient Is Redefining AI Research With A Breakthrough Solution That Surpasses GPT-4o And Perplexity

ChatGPT’s released improved image generation, now available for free

Share link:In this post: ChatGPT’s latest image generator, GPT-4o, is now available for everyone for free. Sam Altman said the free tier would get three image generations daily. Deep Research used to be available only to Plus and Enterprise customers, but it will soon also be available to free users.

OpenAI plans open-weight language model launch after six years

OpenAI will release an open-source weighted language model for the first time since GPT-2 in 2019

OpenAI will release a new “open” AI model soon

Share link:In this post: OpenAI plans to release its first open model since GPT-2 in the upcoming months. OpenAI is seeking community input through a feedback form and will host global developer events soon. Amid competitive pressure, OpenAI CEO Sam Altman pushes for a more open source strategy despite internal disagreements.

How to Start Using AI to Trade Everything from Stocks to Crypto

No one can deny that Artificial intelligence (AI) has transformed many industries, and the trading sector is no exception. For investors looking to profit from cryptocurrencies and stocks, AI does offer an incredible edge, providing tools to profit from complex, fast-moving markets through strategies that adapt in real time. AlgosOne, a top-rated AI-powered trading platform, &hellip; <a href="https://beincrypto.com/how-to-start-using-ai-trade-stocks-crypto/">Continued</a>

OpenAI adds limits after Ghibli AI art surge

$Ghibli Market Cap Hits New High, Studio Ghibli's "Unofficial CEOs" in Panic

OpenAI targets $12.7B revenue as Chinese AI rivals gain

Google launches Gemini 2.5 AI model amid ChatGPT’s 43% market lead

Google rolls out new family of AI reasoning models through Gemini 2.5 models

Share link:In this post: Google recently released a new family of AI reasoning models through its Gemini 2.5 range, beginning with the experimental Gemini 2.5 Pro. The big tech company revealed that the reasoning models will employ “thinking” when answering queries, improving their accuracy and performance. Google also confirmed that the introduction of the new models is part of the company’s plans to make AI smarter and increase its ability to reason.