PWC's OpenAI Deal, AI Misidentification Scandal
30 May 2024: PwC partners with OpenAI, AI wrongly accuses UK woman, EU unveils AI Office, Apple eyes Arc features, US-China AI talks, OpenAI licenses Atlantic, Vox Media
Good Morning,
Latest in AI,
1. 1-Minute Roundup
2. Headlines
3. AI Investment Deals
4. Others1-Minute Roundup
Headlines:
PwC becomes OpenAI's first reseller of ChatGPT Enterprise, aligning with their $1 billion AI investment plan.
A woman in the UK is wrongly accused of shoplifting by AI facial recognition tech, raising accuracy and privacy concerns.
Survey reveals public awareness of generative AI but calls for transparency in AI use in journalism across six countries.
AI Investment Deals:
Ducky raises $2.7M to enhance AI-powered customer service support.
GaiaNet raises $10M to build a decentralised network for open-source AI agents.
Iris.ai raises €7.64M to accelerate AI-assisted scientific research processing.
SmartPM secures $5.5M for its AI-driven schedule controls platform.
Product Fruits raises $1.6M to enhance digital adoption with AI-powered user onboarding workflows.
ExpressionEdits raises $13M for AI therapeutics biotech.
Play AI secures $4.3M for hyper-personalised gaming tools.
Kintsugi AI raises $6M for automated VAT solutions.
ThinkLabs AI raises $5M for AI-powered grid planning enhancements.
Sagetap secures $6.8M for its AI-driven SaaS marketplace.
Rain AI raises $8.1M to develop AI chips.
HoneyNaps raises $11.6M for AI sleep data analysis.
Aisles secures $30M to revolutionise retail with AI.
Exactly.ai raises $4.3M to help artists scale output using AI.
Others:
OpenAI signs licensing deals with The Atlantic and Vox Media.
The EU unveils the AI Office ahead of the AI Act.
Apple may integrate Arc’s AI features into iOS 18, potentially impacting third-party browsers.
US-China AI talks highlight differing visions for AI governance.
Headlines
PwC Becomes OpenAI's First Reseller of ChatGPT Enterprise
PwC signs a deal to become OpenAI's first reseller of ChatGPT Enterprise, the AI firm's business-focused chatbot, making PwC the largest enterprise user. The agreement aligns with PwC's $1 billion AI investment plan.
Why it Matters: The deal marks OpenAI's first reseller agreement as it seeks monetisation avenues. For PwC, it accelerates AI adoption across tax, audit and consulting services.
Numbers:
Over 100,000 PwC employees (75,000 US, 26,000 UK) to get ChatGPT Enterprise licenses
950 of PwC's top 1,000 clients are engaged or planning to use generative AI
Context: Other Big Four firms have also inked major AI deals - KPMG ($2B Microsoft partnership), Deloitte (Google Cloud), and EY (IBM for HR processes).
Between the Lines: OpenAI is loss-making due to AI computing costs, so it turns to enterprise sales like PwC to generate revenue alongside consumer offerings like ChatGPT Plus.Woman Wrongly Accused as Shoplifter by AI Facial Recognition Tech
A woman shares her traumatic experience of being misidentified as a shoplifter by facial recognition system Facewatch used in UK stores, highlighting concerns over accuracy and privacy intrusion as the technology's use grows.
Why it Matters: As facial recognition gains prevalence, scrutiny over its accuracy and guarding against the normalisation of mass surveillance are required to prevent the erosion of civil liberties.Context: Facewatch is used by many UK retailers like Budgens and Sports Direct to identify known shoplifters. Police also use mobile facial recognition vans to scan the public for watchlist matches.
Numbers:
192 arrests in 2024 aided by Met Police facial recognition vans.
The police van camera misidentified 1 in 33,000 people.
Use by Met Police up from 9 times in 2020-22 to 67 times already in 2024.
The police misidentification rate is 1 in 40 facial matches in 2024.
Public Perception of Generative AI Across In News Six Countries
A survey across Argentina, Denmark, France, Japan, the UK, and the USA reveals public awareness of generative AI like ChatGPT but a limited understanding of its use in journalism. Most want disclosure when AI is involved in news production.
Why it Matters: As generative AI capabilities advance, understanding public perceptions is crucial for responsible adoption in sensitive domains like journalism that impact democratic discourse.
Zoom In:
43% think journalists often use AI to edit grammar, and 29% to write headlines.
32% believe human editors check AI outputs before publishing.
AI provides more comfort for soft news than hard news like politics.
50% aware of ChatGPT across six countries.
1-7% use ChatGPT daily (highest in the USA).
56% of 18-24 year olds have used ChatGPT at least once.
Reality Check: The vast majority want at least some disclosure when AI is used to write articles, headlines, analyses, etc. Clear labelling emerges as an expectation as generative AI capabilities advance in the news media sphere.AI Investment Deals
Ducky raises $2.7M in pre-seed funding led by Penny Jar Capital to enhance AI-powered support for customer service teams.
GaiaNet raises $10M in a Series Seed Round to build a decentralised network for open-source AI agents, challenging centralised AI giants.
Iris.ai raises €7.64M led by Silverline Capital and EIC Accelerator Fund to accelerate AI-assisted scientific research processing.
SmartPM, an AI-driven schedule controls platform, raises $5.5M in Series A funding, led by Building Ventures and with participation from GS Futures and The Nemetschek Group.
Product Fruits secures $1.6M in funding from Reflex Capital, bringing total funding to $3.5M. The AI-powered platform creates customisable user onboarding workflows to enhance digital adoption and improve sales targeting.
ExpressionEdits, a protein-based AI therapeutics biotech company, has raised $13M in seed funding led by Octopus Ventures and Redalpine.
Play AI, led by P2 Ventures, Jump Crypto, and others, raised $4.3M in seed funding to build AI tools for hyper-personalised gaming. The platform leverages blockchain and NFTs for data attribution and rewards, enhancing game content and user experiences.
Kintsugi AI secures $6M Series A funding at a $40M valuation to develop automated AI-driven VAT solutions for international sales tax compliance. The platform, founded in 2022, automates sales tax management for over 27M e-commerce and SaaS businesses.
ThinkLabs AI raises $5M in seed funding co-led by Powerhouse Ventures and Active Impact Investments. The NYC-based startup enhances grid planning with AI-powered ThinkLabs Copilot, designed to support ADMS and DERMS and improve edge intelligence.
Sagetap, a San Francisco-based AI-driven marketplace for matching SaaS buyers and sellers, raises $6.8M in funding from NFX, Uncorrelated Ventures, Emergent Ventures, and 15 active customers.
Rain AI, an AI chip startup, raises $8.1M in a Series A Extension round led by Epic Venture Partners. Using in-memory computing technology, the company aims to launch its first chip by 2025, targeting devices like drones, VR goggles, and smartphones.
HoneyNaps, a Seoul-based AI sleep data analysis company, raises $11.6M in Series B funding from Korea Industrial Bank, Hi Investment Partners, and QUAD Investment Management.
Aisles, an AI-driven retail technology company, secured $30M in Series A funding. The company aims to revolutionise the retail shopping experience through advanced navigation, biometric security, and personalised shopping assistance.
Exactly.ai secures $4.3M in seed funding led by Speedinvest to help artists scale their output using AI. The platform allows artists to train AI on their artwork, retaining legal ownership and enhancing production capabilities.
Others
OpenAI Strikes Licensing Deals With The Atlantic and Vox Media
OpenAI has signed licensing agreements with The Atlantic and Vox Media, allowing it to feature its content in products like ChatGPT and use its content to train its AI models.
EU Unveils AI Office Ahead of AI Act Implementation
The EU has detailed the structure of its new AI Office, which will oversee AI governance under the forthcoming AI Act, which will be enforced by July.
The AI Office, effective June 16, will regulate AI risks and promote innovation. It will consist of five units focusing on regulation and compliance, AI safety, testing and evaluation of general-purpose AI models (GPAIs), and providing compliance templates.
Apple Planning to ‘Sherlock’ Arc’s AI Features?
Apple may integrate AI-powered "smart recaps" in iOS 18, potentially mimicking features from Arc, such as AI summaries of web pages and notifications, similar to Arc's "Browse for me" and "pinch to summarise" tools.
This move could reduce demand for third-party browsers like Arc. Still, The Browser Company remains undeterred, emphasising continued innovation in minimising distractions.
US-China AI Talks Highlight Divergent Visions for AI Governance
In a seven-hour meeting in Geneva, US and Chinese officials discussed AI governance, emphasising their desire to prevent catastrophic scenarios while maintaining competitive edges in AI technology.
While the talks did not yield concrete agreements, they set the stage for continued dialogue on AI safety and risks amidst broader geopolitical tensions and efforts to influence global AI governance frameworks.
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