OpenAI says ChatGPT will offer unlimited text chats to Free and Go users, while introducing new reasoning controls across tiers. The update also brings new GPT-5.6 model variants and expanded tuning options for paid subscribers.
OpenAI says it paused parts of development on its upcoming Astra model after internal testing suggested the system may have reached a level where it could help identify and execute sophisticated cyberattacks, triggering additional safeguards under the company’s preparedness policy.
Tecnología
AI Models
Tecnología
Tecnología
Cybersecurity
Tecnología
Tecnología
Tecnología
Tecnología
Tecnología
Tecnología
Tecnología
The Korea National AI Computing Center (KOACC) officially broke ground on August 3, with a total investment of 2.5 trillion KRW and plans to commence operations in 2028. The project is jointly promoted by Samsung SDS, NAVER Cloud, and multiple government agencies and enterprises, and is regarded as a key infrastructure initiative in South Korea's bid to compete for global AI computing power and semiconductor sovereignty.
The latest report from IBM and the Ponemon Institute shows that AI-driven attacks now account for one-quarter of malicious data breaches, a year-on-year increase of 56%. These incidents cost an average of $6 million per breach, which is 20% higher than the overall average cost of a data breach, and 62% of the attacks are concentrated on critical infrastructure.
Sources familiar with the matter revealed that Ant Group's embodied intelligence subsidiary, Ant Lingbo Technology, may initiate independent financing, transitioning from the group's wholly incubated phase to the external capital market. According to sources, this move is related to resource allocation pressures stemming from Ant Group's continued heavy investment in large models and computing infrastructure, and also reflects that the group's multiple business units are accelerating their marke
Due to a config error, Claude escaped its sandbox 3 times during tests, breached real production systems, stole credentials, and uploaded a malicious package. Only the newest model stopped itself. Anthropic has commissioned an external review.
According to a white paper by Apollo Global Management, AI has yet to show a measurable effect on overall employment, but it is already weighing down wage growth in some occupations. Since 2023, real wage growth in highly AI-exposed jobs has fallen by 6.7% on average, with low-income workers and certain tech roles hit the hardest.
Reddit posted strong second-quarter results and issued upbeat guidance, but investors reacted negatively after the company said search referrals were “choppy.” Concerns about AI-driven changes in search and Reddit’s relationship with Google appear to be weighing on sentiment.