•
January 18, 2024
•
7 mins
Hello Niuralogists!
Step into this week's edition as we navigate the dynamic realm of artificial intelligence to present you with the most recent breakthroughs. Our primary focus is to dissect the implications of these updates on various aspects of our lives, ranging from workplaces and businesses to policies and individual experiences. In this issue, we'll unveil compelling advancements, featuring highlights like the spotlight on AI at the Davos World Economic Forum and Microsoft's introduction of the Copilot Pro Subscription with enhanced AI features.
For deeper insights, continue reading...
AI takes center stage at the World Economic Forum in Davos, Switzerland, as prominent figures such as Sam Altman, Satya Nadella, Yann LeCun, and Mustafa Suleyman engage in discussions during the annual gathering of the world's wealthiest and most influential individuals. Notably, OpenAI disclosed its collaboration with the Pentagon on cybersecurity initiatives, marking a shift after lifting a ban on 'military and warfare' applications in ChatGPT's Terms of Service. The creators of ChatGPT are intensifying efforts to combat election disinformation risks, introducing tools to identify DALL-E generated images and implementing additional safeguards to prevent misuse. Altman emphasized the necessity of environmentally friendly power for training billion-parameter models, asserting that future AI advancements hinge on energy innovation. Nadella advocated for a global regulatory approach to AI, expressing optimism about a growing consensus on implementing safeguards. In the broader context of businesses and governments worldwide grappling with AI, this year's forum presents a crucial opportunity for global alignment on adoption and safety.
Microsoft has unveiled Copilot Pro, an upscale subscription offering for their Copilot chatbot that elevates AI capabilities across Microsoft 365 applications. Priced at $20 per month, Copilot Pro provides exclusive features such as AI integration in Office apps and prioritized access to the latest OpenAI models. Available to Microsoft 365 Personal and Family subscribers, this premium subscription includes the introduction of the Copilot GPT Builder. Simultaneously, Copilot for Microsoft 365 is now widely accessible for small businesses, with subscription plans starting at $30 per user per month. Notably, the free version of Copilot, which previously granted free access to GPT-4, now limits such access to "non-peak times." In conjunction with these updates, Microsoft has officially launched the Copilot app on both iOS and Android platforms, as previously disclosed in The Rundown weeks ago. While Microsoft's commitment to advancing AI integration is commendable, there is a degree of disappointment in discovering that GPT-4 will no longer be freely accessible at all times on Copilot.
OpenAI is set to introduce safeguards for its AI tools, particularly the image generation model DALL-E and citations in ChatGPT, as part of an initiative to combat disinformation leading up to the global elections in 2024. The company emphasizes the need for collaborative efforts to protect the integrity of elections and outlines existing safeguards, including a "report" function for users to flag potential violations. OpenAI plans to provide users with real-time news reporting globally through ChatGPT, including attribution and links. Notably, the company commits to implementing image credentials from the Coalition for Content Provenance and Authenticity (C2PA) on its DALL-E 3 imagery, utilizing cryptographic digital watermarking to identify AI-generated content. This move aligns with OpenAI's dedication to truth and accuracy, especially in the context of AI-powered campaigning and impersonations in the political landscape. The company also previews a provenance classifier designed to detect AI-generated images, offering it to a select group of testers for feedback.
Apple is set to close its 121-person artificial intelligence team in San Diego, which focuses on enhancing Siri through data analysis. Employees have been instructed to relocate to Texas by the end of February or face termination on April 26. The team, responsible for assessing Siri voice queries in multiple languages to enhance accuracy, was reportedly taken aback by the abrupt decision, as they had been preparing for an internal move within San Diego. Despite Apple's avoidance of widespread layoffs in the tech industry, this forced relocation has raised concerns and contrasts with the current trend of showcasing brutal layoffs on social media.
Norwegian AI robotics startup 1X Technologies has secured $100 million in new funding, following a $23.5 million Series A round led by OpenAI Startup Fund I LP in March. The company aims to introduce a consumer robot for home use shortly. Their initial android, EVE, currently operates in factories and serves as a security patrol, while the newer NEO model is designed to emulate human movement, assisting with household chores. A tweet from 1X's VP of AI, now deleted, hinted at the imminent availability of NEO models for home purchase. The increasing momentum in the development and competition within the humanoid robot sector suggests that these AI consumer robots may soon become integral in reshaping both home and workplace environments.
📬 Receive our amazing posts straight to your inbox. Get the latest news, company insights, and Niural updates.
The surge in enterprise adoption of ChatGPT, with over 80% of Fortune 500 companies utilizing it, has prompted concerns about the risks of data leaks and intellectual property exposure. To address this, organizations are turning to generative AI-based solutions like Generative AI Isolation to safeguard against potential ChatGPT leaks. Notably, companies such as National Oilwell Varco (NOV) are actively educating their boards on the advantages and risks of ChatGPT, emphasizing ongoing communication and setting expectations. Several technologies, including those from Cisco, Ericom Security by Cradlepoint, Menlo Security, Nightfall AI, Wiz, and Zscaler, are emerging to tackle the security challenge. Ericom Security by Cradlepoint's Generative AI Isolation and Nightfall AI solutions are gaining traction, offering unique approaches to prevent the sharing of confidential data. As the use of generative AI continues to redefine knowledge, CIOs, and CISOs are adopting knowledge-based strategies to pilot and implement AI systems, turning rapid innovation into a competitive advantage while ensuring data protection and compliance.
In preparation for the upcoming 2024 elections, OpenAI has outlined its strategy to address challenges posed by generative AI, focusing on combating misinformation and fake content. The company plans changes to Usage Policies for ChatGPT and APIs, prohibiting applications for political campaigning, lobbying, and those discouraging democratic participation. Authentication efforts include implementing digital credentials from the Coalition for Content Provenance and Authenticity on DALL-E 3-generated images, and enhancing transparency about content origins. OpenAI is experimenting with a tool to detect DALL-E 3-generated images and aims to grant ChatGPT access to real-time global news reporting, including attributions and links, to counter misinformation. While precautions are being developed, caution is advised when consuming election news, emphasizing source verification.
Discover how to experiment with Meta's AudioBox using your voice with these simple steps. Navigate to the AudioBox demo (for U.S. users, a VPN or VPN-enabled browser like Arc may be required) and select 'Try demos.' Opt for "Your Voice" to incorporate your audio sample, or choose from text descriptions or pre-loaded voices. Click 'Record your voice,' where AudioBox will guide you through reading a brief sentence to upload your vocals to the model. Post-recording, input the text you wish to generate, and AudioBox will create two recordings mimicking your vocal style.
.png)