- AI Breakfast
- Posts
- Microsoft Launches Generative AI Certification Program
Microsoft Launches Generative AI Certification Program
AI Breakfast
June 30, 2023
Good morning. It’s Friday, June 30th.
Did you know: 16 years ago today, the original iPhone was released. The now historic keynote presentation by Steve Jobs been watched over 10M times.
In today’s email:
Microsoft Unveils AI Tools for Shopping
Inflection AI Nets $1.3B for AI Assistant
Runway AI Raises $141M for Video Tools
Salesforce Launches Customer Interaction AI Tools
Meta to Develop Larger AI Models
Steam May Reject AI Art Games
China Ramps Up for AI Race
Vatican Co-Releases AI Ethics Handbook
OpenAI Sued Over Privacy Issues
NASA to Use AI in Lunar Communication
Typeface Raises $100M at $1B Valuation
Centaur Labs Gamifies Medical Data Labeling
Databricks Debuts AI Knowledge Engine
Sourcegraph Launches AI Coding Assistant
OpenAI to Open London Office
Microsoft Starts AI Certification Program
10 New AI Tools
Latest AI Research Papers
You read. We listen. Let us know what you think of this edition by replying to this email, or DM us on Twitter.
Today’s edition is brought to you by:
Attending Ai4 this year? We will be!
Join the world's premier AI for business conference at Ai4, uniting the global AI ecosystem through events, media, and resources.
Attend multi-day, networking-enabled conferences featuring the world’s most innovative speakers, attendees, and AI companies.
Today’s trending AI news stories
Microsoft Enhances Online Shopping Experience with New AI Tools on Bing and Edge: Microsoft has unveiled a suite of AI-driven shopping tools on its Bing and Edge platforms, designed to improve the digital shopping experience. The new features, including Buying Guides, Review Summaries, Price Match, and Price Comparison, help users navigate and make informed decisions amidst the influx of options, deals, and customer reviews.
Inflection AI Secures $1.3 Billion Funding to Develop Personal Assistant Pi: AI company Inflection AI has received a hefty $1.3 billion in funding, backed by leading investors Microsoft and NVIDIA. This investment raises the company's total funding to $1.525 billion, which will be used to enhance the development of Pi, its AI-powered personal assistant.
AI Video Startup Runway Attracts $141 Million Investment from Google and Nvidia: Runway AI, a startup creating AI-driven video content, has raised $141 million in funding from major tech companies, including Google, Nvidia, and Salesforce. This investment, which follows a previous funding round of $50 million, increases Runway's total funding to approximately $237 million. Runway plans to utilize the funding to develop and improve its AI-based video generation and editing tools.
Salesforce Introduces Sales GPT and Service GPT to Simplify Customer Interactions: Salesforce has rolled out Sales GPT and Service GPT, AI-driven workflow tools designed to streamline customer interactions for sales and service teams. These tools, powered by Salesforce's AI solution Einstein GPT, aid in closing deals, predicting customer needs, and boosting productivity while prioritizing data security. Salesforce anticipates the integration of generative AI in its other products, including Marketing, Commerce, Slack, Tableau, Flow, and Apex.
Meta Plans to Develop AI Models Significantly Larger than GPT-4: Social media giant Meta has revealed plans to create AI behavior analysis systems for content recommendation that are significantly larger than current models like ChatGPT and GPT-4. However, the specifics remain unclear, and the ambition of building models with tens of trillions of parameters has sparked debate over their necessity and possible implications, especially regarding user privacy and targeted advertising.
Steam Allegedly Rejects Games Using AI Art: Popular PC gaming platform Steam is purportedly refusing games that use AI-generated art, due to potential copyright infringement issues. While not officially confirmed, if true, this policy would significantly affect the use of AI in the gaming industry. The decision has stirred differing opinions among developers and artists, sparking a debate on the practicality of consistently identifying AI-generated assets.
China's Tech Sector Mobilizes for Global AI Competition: China's tech sector is rising to challenge US tech giants like Google and Microsoft in the race for AI dominance. Spearheaded by industry heavyweights and burgeoning startups, China aims to invest an estimated $15 billion in AI this year, with a goal of becoming a major player in the field. However, challenges such as US tech sanctions, data regulations, censorship demands, and limited international expansion for Chinese firms remain significant hurdles.
Vatican Releases AI Ethics Handbook: In a collaborative effort with Santa Clara University, the Vatican has released an AI ethics handbook titled “Ethics in the Age of Disruptive Technologies: An Operational Roadmap.” The handbook seeks to guide tech companies through ethical considerations in AI, machine learning, encryption, and tracking, offering practical advice and actionable steps before government regulations are established.
OpenAI Faces Lawsuit Over Alleged Privacy Violations: ChatGPT creator OpenAI is facing a lawsuit alleging the theft of personal data and violation of privacy. The plaintiff alleges that OpenAI scraped 300 billion words from the internet, including personal data, without consent, risking potential "civilizational collapse." The lawsuit, which seeks class-action status, accuses OpenAI of prioritizing profit over privacy laws. OpenAI has yet to respond to the allegations.
NASA to Utilize AI Similar to ChatGPT for Lunar Gateway Station Communication: NASA plans to use an AI system similar to ChatGPT for its lunar Gateway space station to facilitate astronaut communication and help navigate complex systems and monitor mission parameters. The AI system will also provide support for future Artemis missions and potential spacecraft issue detection. NASA emphasizes the crucial role of AI in space exploration, particularly in the Lunar Gateway project.
AI Startup Typeface Raises $100M at a $1B Valuation: Typeface, an AI startup founded by former Adobe CTO Abhay Parasnis, has raised $100 million in a Series B funding round, led by Salesforce Ventures. The funding will be used to expand Typeface's platform and team. The startup, which focuses on delivering personalized content for brands at scale using generative AI, offers a content hub, AI-powered training and personalization, and templates and workflows for integration into existing systems.
Medical Data Labeling Gamified to Advance AI: Centaur Labs, an MIT-founded startup, is transforming medical data labeling through gamification. Their app, DiagnosUs, allows medical experts to review and classify real-world scientific and biomedical data, earning small cash prizes for accurate classifications. The data collected helps train and refine AI algorithms used in biotech, pharmaceutical development, and medical device commercialization.
Databricks Introduces AI-Powered Knowledge Engine LakehouseIQ for Businesses: Databricks has unveiled LakehouseIQ, an AI-powered knowledge engine designed to understand an organization’s data environment. Drawing on insights from across the Databricks Lakehouse platform, LakehouseIQ aids in the development of custom AI models and tools. It features an AI assistant for generating and explaining queries, improved search functionality, and offers insights for datasets and workflows.
Sourcegraph Releases AI Coding Assistant Cody 5.1: Sourcegraph has launched Cody version 5.1, an upgraded version of its AI coding assistant, capable of writing entire files and tests, fixing bugs, and refactoring projects with minimal human intervention. Cody 5.1 boasts an expanded understanding of code context across repositories and enhanced automation capabilities. Cody 5.1 is available for free to developers working on public and private code.
OpenAI Challenges DeepMind with New London Office: OpenAI has announced its intention to open its first international office in London, in a move that brings it into direct competition with Google’s AI research division, DeepMind. The London office will allow OpenAI to advance research and engineering capabilities, while fostering collaboration with local communities and policymakers. The decision to choose London is strategic, given the city's reputation as a hub for AI startups.
Microsoft Launches Generative AI Certification Program: Microsoft has initiated an AI Skills Initiative, offering free generative AI courses via LinkedIn as part of its Skills for Jobs program. The courses cover introductory AI concepts and responsible AI frameworks, with certification upon completion. This initiative seeks to equip learners with the necessary skills to cope with the increasing integration of generative AI into the workforce. The course will initially be available in English, with plans to expand to other languages in the future. Apply here by August 15th to enroll.
🎧 Did you know AI Breakfast has a podcast read by a human? Join AI Breakfast team member Luke (an actual AI researcher!) as he breaks down the week’s AI news, tools, and research: Listen here
10 new AI-powered tools from around the web
arXiv is a free online library where scientists share their research papers before they are published. Here are the top AI papers for today.
Microsoft introduces a framework that utilizes programmatic supervision to calibrate LLMs and detect eros. By aligning LLM outputs with other supervision sources through a harmonizer model, risk scores are assigned to uncertain responses. Experimental results showcase improved accuracy and surpass advanced models on challenging evaluation datasets. Notably, this approach does not rely on manual efforts or labeled training data, making it a versatile solution applicable in various domains.
Researchers conducted a study to evaluate GPT-3.5 & 4, and compared their performance with human tutors in various programming scenarios. They used real-world Python programming problems and assessed the models’ outputs using expert evaluations. The results showed that GPT-4 performed exceptionally well and came close to human tutors’ performance in several scenarios. However, there are still areas where improvement is needed.
Researchers introduce OBELISC, a publicly available dataset comprising 141 million web pages extracted from Common Crawl. The dataset consists of interleaved image-text documents, containing 353 million associated images and 115 billion text tokens. The researchers describe the dataset creation process, including comprehensive filtering rules, and provide insights into its content. They demonstrate the dataset’s viability by training an 80 billion parameter vision and language model on OBELISC, and achieving competitive performance on various multimodal benchmarks.
Alibaba Group’s DAMO Academy introduces 3D-Speaker, an expansive corpus for speech representation disentanglement. With over 10,000 speakers recorded simultaneously by multiple devices, at different distances, and speaking various dialects, this groundbreaking resource offers opportunities to untangle complex speech representations. It facilitates research in supervised and unsupervised methods, out-of-domain learning, and the evaluation of universal speech models.
LENS (Large Language Models ENhanced to See) is a modular approach that combines LLMs with independent vision modules to tackle computer vision tasks. Unlike previous methods, LENS eliminates the need for multimodal pretraining and extensive data, making it computationally efficient. By leveraging pretrained vision modules and a frozen LLM, LENS achieves competitive performance in zero-shot object recognition and vision-language tasks. It can be applied to any off-the-shelf LLM and offers flexibility and scalability.
Thank you for reading today’s edition.
Your feedback is valuable.
Respond to this email and tell us how you think we could add more value to this newsletter.
🖇️