AI Safety Chiefs Resign, Microsoft's China Move, Sony Music Fights AI Theft
17 May 2024: OpenAI's leadership shake-up, Microsoft moves AI staff, Sony battles AI copyright infringement, Reddit partners with OpenAI, Snowflake $1B Reka AI deal
Good Morning & Happy Weekend,
Latest in AI,
1. 1-Minute Roundup
2. Headlines
3. AI Investment Deals
4. Other AI News1-Minute Roundup
Headlines:
OpenAI’s Key Safety Chiefs Resign: Jan Leike and Ilya Sutskever leave, causing uncertainty in AI safety efforts.
Microsoft Relocates China-Based AI Staff: Hundreds of AI employees asked to move due to U.S.-China tech tensions.
Sony Music Warns AI Firms on Content Use: Unauthorised AI use of music content sparks industry clash.
AI Investment Deals:
Laws of Motion Raises $5M: NYC-based e-commerce AI sizing tech provider secures seed funding.
Gamma Secures $12M: AI-powered presentation platform completes Series A funding.
Firmus Funding Round: Miami-based AI pre-construction firm raises an undisclosed amount.
Recall.ai Raises $10M: AI meeting data API startup reaches Series A milestone.
Voxel51 Secures $30M: Visual AI company expands with Series B funding.
Leya Raises $10.5M: Swedish AI legal assistant startup completes seed round.
Angel AI Secures $4.75M: AI platform for children raises seed funding.
ProjectDiscovery Raises $25M: AI cloud security platform completes Series A round.
Snowflake in Talks to Acquire Reka AI: Potential $1 billion acquisition of language model startup.
Others:
Reddit Inks Data Licensing Deal with OpenAI: Partnership integrates Reddit content into ChatGPT.
Arup Loses $25M in Deepfake Scam: British firm defrauded via AI deepfake technology.
OSI seeks public input on "Open-source AI": Workshops to define open-source standards for AI.
Wiley Shuts Down 19 Journals: AI-driven publishing fraud leads to massive journal closure.
Hugging Face Commits $10M in Free Compute: Initiative to democratise AI development for small players.
Headlines
OpenAI's Key Safety Chiefs Resign
OpenAI's Jan Leike and Ilya Sutskever resigned, leaving the team responsible for ensuring AI safety from superintelligence without leadership.
Why it Matters The resignation of key leaders jeopardises OpenAI’s mission to ensure AI safety and alignment with human values.
Zoom In:
Jan Leike and Ilya Sutskever led the superalignment team.
Their team aimed to control superintelligent AI, ensuring it aligns with human intent.
OpenAI admits no current solution exists for controlling superintelligent AI.
Backstory:
The superalignment team was formed to address AI safety, dedicating 20% of OpenAI’s compute resources.
Leike joined OpenAI from Google's DeepMind.
The superalignment team was given 20% of OpenAI’s compute power.
OpenAI aimed to have a breakthrough within four years to control superintelligent AI.
Zoom Out:
Resignations highlight challenges in AI safety and alignment.
OpenAI must find new leaders to continue critical safety work.
The future of AI safety efforts remains uncertain amid leadership changes.
Microsoft Relocates China-Based AI Staff
Microsoft asks hundreds of China-based AI employees to consider relocating due to escalating U.S.-China tensions and stricter controls on technology access.
Why it Matters The move reflects escalating U.S.-China tech rivalry and tighter U.S. controls on technology exports to China.
Zoom In:
Microsoft offers 700-800 China-based engineers relocation options.
Microsoft employs about 7,000 engineers in Asia-Pacific, mainly in China.
Destinations include the U.S., Ireland, Australia, and New Zealand.
Engineers can stay in China, but relocation supports global cloud ambitions.
Context: The Biden administration tightens AI technology access for China, impacting U.S. companies with Chinese operations.
Big Picture:
Relocation may hamper China's AI talent pool.
Microsoft maintains a significant presence in China despite tensions.
U.S.-China tech conflict intensifies, affecting global tech strategies.
Sony Music Warns AI Firms on Content Use
Sony Music has warned AI companies to stop unauthorised use of its artists' content, emphasising the need for control and compensation.
Why it Matters: Unauthorised AI use of copyrighted content undermines artists' rights and potential earnings, escalating tensions between tech and creative industries.
Zoom In:
Sony Music sent letters to 700+ companies.
Content includes album art, metadata, music compositions, and lyrics.
Unauthorised use affects control and compensation for artists and the company.
Universal Music Group's dispute with TikTok lasted over a year.
Two voiceover actors filed a class action lawsuit against Lovo Inc.
Zoom Out:
AI-driven copyright infringement is a growing issue.
Universal Music Group previously pulled its catalogue from TikTok to protect rights.
US legislation, like the "NO FAKES" Act, aims to safeguard artists’ voices and images from unauthorised AI use.
AI Investment Deals
Laws of Motion, an NYC-based AI sizing tech provider for eCommerce brands and retailers, raised $5M in seed funding from investors, including Corazon Capital and Sequoia Capital.
Gamma, a San Francisco-based AI-powered presentation platform, raised $12M in Series A funding led by Accel with participation from Script Capital and others.
Firmus, a Miami-based AI pre-construction design review, bid-ability, constructability review and risk analysis company, raised an undisclosed amount in funding led by Navitas Capital.
Recall.ai, a YC-backed developer AI-powered meeting data API infrastructure startup, raised $10M in Series A funding led by Ridge Ventures, bringing its total raised to over $12M.
Voxel51, an Ann Arbor-based visual AI company, raised $30M in Series B funding led by Bessemer Venture Partners.
Leya, a Swedish AI legal assistant startup, raised $10.5M in seed funding, led by Benchmark's Chetan Puttagunta and with Hummingbird, SV Angel, and Y-Combinator participation.
Angel AI, a Nashville-based ChatGPT-like AI platform for children aged 5-12, raised $4.75M in seed funding led by Cortical Ventures.
ProjectDiscovery, a California-based AI-powered cloud security platform, raised $25M in Series A funding led by CRV with participation from SignalFire, Point72, and others.
Snowflake Inc. is in talks to acquire Reka AI, a startup founded by Google and Meta researchers, for over $1 billion. Reka AI specialised in large language models and was valued at $300M in 2023.
Others
Reddit Inks Data Licensing Deal with OpenAI, Adds AI Partner
Reddit has signed a deal giving OpenAI access to its data to integrate Reddit content into ChatGPT and new AI products.
OpenAI will become a Reddit advertising partner as part of the deal, and Reddit users and moderators will gain new AI-powered features.
British Engineering Company Arup Loses $25 Million in Deepfake Scam
Fraudsters used a digitally cloned CFO to trick Arup staff into transferring $25 million.
Hong Kong police continue investigating the case, highlighting the growing threat of AI deepfake technology in cybercrime.
OSI Seeks Public Input to Define "Open Source AI"
OSI launches a global workshop series to solicit feedback on its Open Source AI Definition draft (v0.0.8).
Workshops address whether existing open-source licenses adequately cover AI models and datasets.
Wiley Shuts Down 19 Journals Amid AI-Driven Publishing Fraud
Publisher Wiley has discontinued 19 scientific journals overseen by its Hindawi subsidiary due to systematic manipulation by "paper mills."
Over 11,300 papers have been retracted from Wiley's Hindawi portfolio in the last two years linked to unethical practices like AI manuscript fabrication.
HuggingFace Commits $10M in Free Compute to Democratise AI Development.
Hugging Face offers $10 million in accessible shared GPUs through its ZeroGPU program to support small developers, academics, and startups in AI innovation.
CEO Clem Delangue emphasises the importance of decentralising AI advancements, making state-of-the-art technologies accessible to a broader community and countering the dominance of tech giants.
Like TheAIML?
Please don’t keep it to yourself. Colleagues and friends can sign up here.
For even more,
Our newsletter archive: Click here
Follow us on Linkedin,
Follow us on 𝕏/Twitter
Your feedback is crucial in helping us improve and deliver the content you want to see. Please Email us