$HACHI Nedir?

On August 20, JPMorgan stated that following the announcement of a 40 trillion won ($29 billion) stock buyback plan, SK Hynix may return at least $130 billion to shareholders by 2027. JPMorgan analyst Jay Kwon noted in a report that when SK Hynix announced its stock buyback plan on Wednesday, it raised its commitment to shareholder returns for the period from 2025 to 2027 to over 50% of cumulative free cash flow, up from a previous cap of 50%. Kwon estimates that this means the company will provide at least an additional 180 trillion won in shareholder returns by 2027, which is equivalent to 16% of its current total market value, beyond the already announced plans. He believes this will help support the stock price following recent sell-offs. Kwon stated that the worst period is over and expects market sentiment towards SK Hynix to gradually improve in the medium term, recommending investors to increase their holdings.

On August 20, JPMorgan stated that following the announcement of a 40 trillion won ($29 billion) stock buyback plan, SK Hynix may return at least $130 billion to shareholders by 2027. JPMorgan analyst Jay Kwon noted in a report that when SK Hynix announced its stock buyback plan on Wednesday, it raised its shareholder return commitment for the period from 2025 to 2027 to over 50% of cumulative free cash flow, up from a previous cap of 50%. Kwon estimates that this means the company will provide at least an additional 180 trillion won in shareholder returns by 2027, equivalent to 16% of its current total market value, beyond the already announced plans. He believes this will help support the stock price following recent sell-offs. Kwon stated that the worst period is over and expects market sentiment towards SK Hynix to gradually improve in the medium term, recommending that investors increase their holdings.

According to Beating AI news, OpenAI has previewed a new service called Private Safety Processing, which allows eligible enterprises and API customers to maintain zero data retention while using advanced models. The system can check for potential abuse across multiple rounds of dialogue, but OpenAI employees cannot see the customer's prompts or model responses. Customer data can remain on their own servers or be stored with OpenAI, but encrypted with the customer's own keys, which OpenAI cannot access. If the system detects risks, it will only return a limited safety signal to OpenAI, such as identifying the type of suspicious activity, without sharing the original dialogue. This directly addresses a controversial point raised by Anthropic. Fable 5 and Mythos 5 require retention of user inputs and model outputs for 30 days to monitor for attack behaviors across multiple requests. Microsoft previously restricted employee use of Fable 5 due to this policy. Anthropic believes that data retention is a necessary cost for effective security monitoring, while OpenAI aims to prove that both objectives can be achieved simultaneously. Private Safety Processing is currently being tested with a small number of customers, and OpenAI plans to launch it in September along with a technical white paper. Regular ChatGPT Free, Plus, Go, and Pro users are not included in this ZDR scheme, and existing data settings will not change.