- Front Research
- Posts
- OpenAI Enhances ChatGPT with Improved Directness and Language Skills
OpenAI Enhances ChatGPT with Improved Directness and Language Skills
+ 6 break tech news stories
TECH BRIEF
Meta has started testing its AI chatbot, Meta AI, with WhatsApp, Instagram, and Messenger users in India and parts of Africa to tap into its massive user bases and enhance its AI services. This development comes in response to big moves from other tech giants like OpenAI. India is Meta's largest single market, housing over 500 million Facebook and WhatsApp users. With AI now a key user and investor expectation, Meta is using Instagram and WhatsApp's global reach to potentially outdo rivals in AI scale and accessibility.
Roku has revealed a second data breach affecting 576,000 users, identified while investigating the previous attack that compromised 15,000 accounts. The latest breach involved the technique called "credential stuffing", where data from other sources was used to hack into Roku accounts. In fewer than 400 cases, unauthorized purchases were made. The company has reset passwords for affected accounts and is implementing two-factor authentication.
OpenAI CEO, Sam Altman, has been pitching AI services, including ChatGPT Enterprise, to Fortune 500 company executives. Despite countering backer, Microsoft, the demos aim to drum up potential revenue from businesses across various industries. OpenAI projects a $1 billion target revenue for 2024 and expects enterprise sales to play a significant role in this, with over 600,000 signups for ChatGPT Enterprise so far.
OpenAI has upgraded its AI-powered chatbot, ChatGPT, for premium users featuring an improved and updated version of GPT-4 Turbo, capable of refined writing, math, logical reasoning, and coding. The new model is designed to provide more direct and less verbose responses with updated conversational language. This enhancement follows an unflattering week for the company involving alleged information leaks and military pitches.
In Q1 2024, Carta's initial analysis reveals rising valuations for most venture tiers, with the exception of late-stage Series D and E+. Round sizes saw an upward trend - seed, Series A to C experienced robust growth, particularly in priced seed and Series C stages. Final figures on rounds and capital raised are still pending, but expectations point to continuous modest growth in VC activity.
Google has developed a new technique to provide limitless context to Language Models (LLMs), offering an improved understanding of languages and contexts. The groundbreaking method promises to significantly boost the capabilities of AI language understanding.
Reply