OpenAI, the company behind ChatGPT, has raised $6.6 billion in new funding, pushing its valuation to an estimated $157 billion. The funding round saw participation from major investors such as Microsoft, Nvidia, Thrive Capital, and Khosla Ventures. Despite recent restructuring and the sudden exit of longtime Chief Technology Officer Mira Murati, investor confidence remains high, with many believing in the company’s strong growth potential. Thrive Capital alone has committed $1.2 billion and may invest another $1 billion next year if revenue targets are met.
OpenAI is in the midst of restructuring, moving away from its non-profit origins towards a more commercial, for-profit model. The recent funding could convert into equity if this transition succeeds. CFO Sarah Friar suggested a potential buyback of employee shares, though no concrete plans have been set. Investors have also secured protections, allowing them to renegotiate the valuation if the restructuring is not finalised within two years.
Since launching ChatGPT, OpenAI has seen rapid growth, attracting 250 million weekly active users. Despite incurring heavy losses, the company anticipates generating $3.6 billion in revenue this year, with projections reaching $11.6 billion in 2024. As it scales, OpenAI remains committed to its pursuit of artificial general intelligence (AGI), aiming to advance AI capabilities while moving towards profitability.
Google’s Gemini Nano, a powerful on-device AI model, is now available for developers to integrate into their apps through the newly released AI Edge SDK. By running locally, Gemini Nano offers tasks such as text summarisation and image descriptions, while keeping user data private by processing everything on the device.
Already featured in Google’s Pixel 9 and Samsung’s Galaxy S24 devices, Gemini Nano powers AI functionalities in apps like Pixel Recorder and Google Messages. Developers can now experiment with these tools to bring AI features to their own apps, with Google expanding access to the AI Edge SDK beyond its previous early access programme.
Currently, developers can explore text-to-text prompts, such as smart replies, proofreading, and summarisation. Google plans to add support for other modalities, like image processing, in future updates. This move will enable broader AI integration across third-party apps, offering enhanced user experiences.
By customising Gemini Nano through the AI Edge SDK, developers will have control over how AI processes information, allowing them to adapt responses to suit their app’s needs. This marks a significant step towards more AI-driven apps for Android users.
Microsoft has officially launched the OpenAI library for .NET, offering comprehensive support for OpenAI’s REST API and flagship models like GPT-4.0. Designed to simplify integration for developers, the library enables the use of OpenAI and Azure OpenAI services within .NET applications.
Following a beta release in June, the stable version is now available through NuGet. It includes full support for models such as GPT-4.0 mini and o1-preview, while providing flexibility for developers to create extensions and additional libraries for specific needs.
The library also includes both synchronous and asynchronous APIs, allowing developers to choose between different patterns for their applications. Other key features include streaming completions for more dynamic interactions, and compatibility with .NET Standard 2.0, ensuring broad usage across different platforms.
This open-source library, available on GitHub, complements OpenAI’s existing libraries for Python and JavaScript, making it easier for developers to work with OpenAI technologies in .NET environments.
An AI-powered camera system has been introduced on Tavistock Road, Plymouth, to detect road traffic offences. The technology captures images of passing vehicles, checking for seatbelt use and drivers using mobile phones. While the AI initially identifies potential offences, the final decision is made by human reviewers. Offenders may receive a warning letter or prosecution notice.
Adrian Leisk from Devon and Cornwall Police emphasised the safety risks associated with not wearing seatbelts and using mobile phones while driving. He highlighted several recent fatal incidents linked to these offences. Authorities aim to encourage safer driving habits rather than penalise motorists.
Plymouth City Council revealed that similar AI camera deployments in 2023 along other roads in Devon and Cornwall showed positive results. Data from the A30 and A38 reported low offence rates, with 0.31% for mobile phone use and 0.38% for seatbelt violations.
Authorities hope that the new system will continue to reduce driving offences and improve road safety. The initiative focuses on changing driver behaviour, with the ultimate goal of preventing accidents caused by distractions and failure to use seatbelts.
Jensen Huang, Nvidia’s CEO, has described demand for the company’s AI chips as ‘insane’, reflecting the increasing global interest in AI technology. His remarks came as Nvidia announced an expanded partnership with IT consultancy Accenture, aimed at scaling AI solutions for businesses worldwide.
The collaboration will see a new business group formed, focused on building custom AI systems using Nvidia’s cutting-edge technology. The partnership also involves Meta’s open-source AI models, Llama, further reinforcing Nvidia’s position as a major player in the growing AI ecosystem. Huang highlighted the role of the partnership in addressing global AI demand, marking the start of what he termed the ‘enterprise AI’ wave.
As corporations scramble to build AI infrastructure, Nvidia’s dominance in AI hardware, particularly in graphics processing units (GPUs), has been a key driver of the company’s success. Nvidia’s stock has surged, closing 1.6% higher, and more than doubling in value this year, while Accenture’s shares also rose by 1.2%.
Nvidia’s success is driven by widespread adoption of AI across industries such as healthcare, cloud computing, and finance. The partnership with Accenture represents the latest step in Nvidia’s strategy to secure its leadership in the enterprise AI market, which is poised for exponential growth in the coming years.
OpenAI has launched several new tools aimed at making it easier for developers to create applications powered by its AI technology. Among the key innovations is a real-time tool that allows developers to build AI voice applications using a single set of instructions, streamlining what was previously a multi-step process.
The startup, supported by Microsoft, also introduced a fine-tuning tool that enables developers to improve AI model responses using both text and images. This enhancement boosts capabilities like visual search and object detection, potentially benefiting sectors such as autonomous vehicles.
OpenAI has forecast a rapid rise in revenue, expecting to generate $11.6 billion next year, driven by businesses building their own AI apps using its technology. With competition from tech giants like Google heating up, OpenAI is focused on rolling out advanced tools to retain its edge in the generative AI race.
Other newly unveiled features include a method for smaller AI models to learn from larger ones, and a ‘Prompt Caching’ system that can reduce development costs by reusing previously processed text, cutting expenses by up to half.
A team of researchers from the Robot Learning Lab in London has developed an innovative way to train robots using AI-generated images. The system, named Genima, fine-tunes Stable Diffusion to map out robot movements, guiding them in both virtual and real-world environments. This research is set to be presented at the Conference on Robot Learning next month.
Genima aims to improve robots’ ability to complete tasks, such as picking up objects or folding laundry. The system uses images as both input and output, helping robots better understand the tasks they’re performing and reducing errors, like moving into walls. It could revolutionise training for a wide range of robots, from mechanical arms to driverless cars.
Researchers successfully tested Genima on 25 simulated and nine real-world tasks, with average success rates of 50% and 64% respectively. While these numbers aren’t perfect, the team is optimistic that the use of video-generation AI models could boost speed and accuracy, making future applications more efficient.
The versatility of Genima is promising, with the potential to be applied to many different kinds of robots. Its ability to use image data for decision-making could lead to smarter, more capable machines in everyday life and industry.
President Joe Biden has signed legislation that will exempt certain United States semiconductor manufacturing facilities from additional federal environmental reviews, helping to accelerate projects funded by the $52.7 billion CHIPS Act. The move is aimed at preventing potential delays that could arise from lengthy environmental assessments required under the National Environmental Policy Act.
While proponents argue that these projects have already complied with various environmental regulations at federal, state, and local levels, environmental groups like the Sierra Club caution that the reviews are essential to protect communities and workers from hazardous materials used in chip production. Critics are concerned about the risks of bypassing such safeguards.
The legislation is seen as a critical step to bolster the US semiconductor industry, with companies like Samsung, Intel, and Taiwan‘s TSMC set to benefit from billions in government subsidies. These funds are intended to strengthen supply chains, create jobs, and reduce dependence on foreign suppliers like China.
Despite the bipartisan support, some lawmakers, including Representative Zoe Lofgren, voiced opposition, citing past instances of semiconductor-related pollution. Lofgren argued that the reviews are a necessary tool to prevent similar environmental harm in the future.
The US Commerce Department announced its plan to allocate $100 million to promote the use of AI in developing sustainable semiconductor materials. This funding initiative is part of a broader effort overseeing $52.7 billion designated for US chip manufacturing and research, aimed at strengthening the country’s position in the semiconductor industry.
The new funding will support universities, national laboratories, and private sector companies in creating AI-driven autonomous experimentation methods. By harnessing the capabilities of AI, the initiative seeks to streamline and expedite the development of innovative semiconductor materials that are less resource-intensive, ultimately contributing to a more sustainable manufacturing process.
With the semiconductor industry facing increasing pressure to reduce environmental impact, this investment represents a significant step towards integrating advanced technologies to foster sustainable practices. The Commerce Department’s focus on AI in this sector underscores the potential for transformative advancements that can meet both economic and environmental goals, helping to secure a more resilient supply chain for the future.
Global financial messaging service SWIFT will trial live transactions of tokenised assets and digital currencies in 2024, aiming to accelerate their integration into the financial system. Tokenisation, which transforms traditional assets like bonds into digital units, promises faster, cheaper, and more efficient trading by cutting out intermediaries.
Despite high expectations, tokenisation and digital currencies have yet to achieve widespread adoption. Around 90% of central banks are experimenting with digital currencies, hoping to modernise trade and payments in the evolving cryptocurrency landscape. SWIFT has already tested Central Bank Digital Currencies (CBDCs) and plans to connect them with existing financial infrastructure.
SWIFT’s head of innovation, Nick Kerigan, stated that demand is growing for real-world digital asset transactions where payment in real money happens simultaneously. However, market fragmentation has limited progress, with most initiatives still confined to banks’ internal systems.
The latest SWIFT trials will involve trading various digital assets across multiple platforms. Kerigan emphasised the need for both delivery and payment in tokenised transactions, highlighting the role of wholesale CBDCs and tokenised deposits in making this possible.
