China Plans Regulations on AI: Musk, Tipalti Integrates GPT-4 for Increased Efficiency and more

Pivot 5: 5 stories. 5 minutes a day. 5 days a week.

1. China to Initiate AI Regulations: Elon Musk

Billionaire Elon Musk indicated China's plans to introduce artificial intelligence (AI) regulations following his discussions with senior Chinese officials during a recent visit.

Musk, CEO of Tesla Inc, and owner of Twitter, stated that these productive talks centered on the risks of AI and the necessity of oversight and regulation. Earlier in April, China's cyberspace regulator proposed draft rules for managing generative AI services, including a requirement for firms to submit security assessments prior to public launch. The Cyberspace Administration of China (CAC) has supported AI innovation and application, given that content generated aligns with the nation's core socialist values.

They also stressed that providers will be responsible for data legitimacy used to train generative AI products, advocating for preventative measures against discrimination in algorithm design and data training.

Read the full story here

2. Tipalti Embeds OpenAI's GPT-4 into Payables Platform, Enhancing Efficiency

Tipalti, a global leader in payables automation, has integrated OpenAI's GPT-4 technology into their payables intelligence engine, Tipalti Pi.

This enhancement aims to increase productivity through improved Auto Coding and Ask Pi capabilities. By harnessing generative AI, Auto Coding can more accurately predict invoice and purchase order coding, eliminating manual work and improving spend visibility. In internal tests, this saved up to 40% of additional processing time. Ask Pi, an AI-powered digital assistant, expedites information retrieval and offers quick answers to key business queries. Both Auto Coding and Ask Pi are in beta testing, with general availability for all customers slated for Q3 2023.

With these advancements, Tipalti aims to revolutionize the payables process, empowering finance professionals with valuable insights and enabling focus on strategic decision-making.

Read the full story here

3. Media Outlets Rally for AI Content Usage Rules

In light of the rise of generative AI programs such as OpenAI's ChatGPT and Google's Bard, media organizations including The New York Times and NBC News are discussing the establishment of regulations for AI usage of their content.

This move is in response to concerns that AI could undermine their business models by republishing content without credit, or creating misinformation, thereby eroding online news trust. Digital Content Next, an industry group, has released seven principles for the governance of generative AI, addressing safety, compensation for intellectual property, transparency, accountability, and fairness. The group also suggests that deployers of generative AI systems should be held accountable for system outputs. These principles are intended to prompt further discussion rather than act as definitive industry rules.

The coming months could prove challenging as media organizations grapple with the potential disruptions caused by AI.

Read the full story here

4. 8Flow Raises $6.6M to Automate Customer Support Tasks

8Flow.ai, an automation startup, has launched with a $6.6 million seed funding round.

The company is developing a self-learning workflow automation engine aimed at streamlining the tasks of customer support agents. The engine integrates with popular tools like Zendesk, ServiceNow, and Salesforce Service Cloud. The next phase involves using gathered data to train machine-learning models for AI-driven workflows customised to individual user needs. 8Flow's current tool is a Chrome extension that automatically learns common steps for each agent, reducing the need for repetitive tasks such as copying and pasting data across applications. The team, led by co-founder and CEO Boaz Hecht, has experience in workflow automation, with Hecht having previously co-founded and sold SkyGiraffe, an enterprise mobility platform.

The seed round was led by Caffeinated Capital and included participation from BoxGroup, Liquid2, HNVR, Trilogy, and several angel investors.

Read the full story here

5. SentiLink Turns Synthetic ID Fraud into a Booming Business

Hasan Hakim Brown manufactured synthetic identities using real social security numbers and fictitious names to defraud the Small Business Administration and various banks out of more than $20 million.

SentiLink, a San Francisco-based startup, spotted this pattern and turned the insight into a thriving business. Founded by Naftali Harris and Maxwell Blumenfeld, SentiLink uses AI, along with manual review, to identify potential fraudulent activity. The company raised $70 million in 2021, reaching a valuation of $430 million. It now serves seven of the U.S.’ 15 biggest banks and six of the 10 largest credit unions, among its 300-plus customers.

Despite the success, SentiLink holds only a small portion of the $15 billion fraud-prevention market, facing competition from companies like Experian, Lexis-Nexis, and Socure.

Read the full story here