All Things Data: Women in AI, AI Fairness, Conversational AI and AI Metrics

The AI fairness debate continues among enterprises, tech leaders and in 2021 companies are focusing on AI literacy to drive change. How can we build ethical, transparent and fair AI systems that reflect our values? Everything begins with education and keeping stakeholders in the loop about technology.

With artificial intelligence systems applied across different sectors, literacy on AI fairness should follow suit because technology is moving fast leaving organizations without knowledge on implementing responsible AI systems1. Trust in AI systems requires communication and building responsible technology that reflects our needs.??

Voice technology is evolving with more startups making forays in the voice market in 2021. Venture funding is trickling in for voice technology startups with Ellipsis Health using machine learning and deep learning to analyze voice patterns to detect depression. Depression still poses problems to society and the work of Ellipsis Health highlights the potential of startups in health and voice spaces. Other startups in health tech working on exciting solutions include Kintsugi, Sonde Health and Vocalis Health.

Customer engagement is vital for every business and conversational intelligence is evolving as companies look to optimize customer interactions. Companies face hurdles gaining insights from customer data across multiple channels including email, chat and social media but this is changing. The acquisition of ScopeAI by Oberseve.AI, a startup that analyzes call center data attests to the gradual change towards using technology to improve customer engagement.?

The Omni channel analytics space is ripe for growth as Observe.AI announced plans to integrate text and voice in a single dashboard for mapping the customer journey allowing companies to manage the customer experience on a single platform.

Female tech leaders are championing AI ethics with notable names including Noelle Silver, Nuria Oliver, Briana Brownell and Arezou Soltani among others. These AI champions are changing the narrative on AI ethics with efforts including research papers, building successful startups and securing patents for their work.

How can AI technology work in the interests of humans? What questions should organizations be asking on AI ethics? In addition, what are companies doing to ensure adoption of responsible and ethical technology? These female tech leaders in AI ethics ask these questions as they educate society on harnessing the power of technology for benefit to society.

Learn about these and more tech bytes in our weekly AI update.

Educating Enterprises on AI Fairness

AI adoption among organizations rose in 2021 in response to the pandemic with companies adjusting to remote operations and relying on technologies such as Zoom and Slack to coordinate remote staff. Post pandemic, organizations have realized the value of becoming data driven with more investments in technologies including machine learning, 5G, IOT and deep learning.

Despite this technological adoption, the question of trust in AI systems remains unresolved. As AI powers technologies in this digital economy, trust is vital to ensure that algorithms are fair and not biased. No one wants biased algorithms2 determining their credit score in a bank or unfair ML algorithms recommending their shopping products.

Building this trust begins with educating organizations about AI fairness. Literacy on the fairness of AI systems is critical for every company that wants to use technology to add value to customers and scale. The executive leadership should support AI fairness efforts alongside data scientists and data engineers to foster trust in technology.

Sentiment Analysis and Brand Strategy

Every brand understands the value of sentiment analysis3 in measuring the favorability of their product in the market and customer perceptions. However, brands miss important details about their customers because of less effective data analysis methods but with machine learning transforming business marketing, the brand experience for companies is changing because of better algorithms capable of sorting vast data and offering accurate insights.

The recent algorithms out of University of Maryland will help brands leverage machine learning for sentiment analysis and understanding relationships between customers and brands. This algorithm can review good and bad comments regarding the brand on social media platforms and offer insights about the positioning of the brand in the market.

Women Championing AI Ethics in 2021

The debate on responsible and ethical AI? has been raging for years now and in 2021, the message continues with women joining the cause. From research work and creating successful startups, women including Noelle Silver, Nuria Oliver, Briana Brownell and Arezou Soltani are making this happen.?

A recent guest on the HumAIn Podcast, Noelle Silver is an excellent example of women advocating for responsible use of artificial intelligence. She is dedicated to educating companies on AI and helping them ask the right questions while supporting minority women to understand technology. Nuria Oliver is a woman veteran in the AI industry with her work on developing technology that works for humans and not against them. With over 100 research papers, Nuria advocates for positive AI influence in society.

Technology Investments and Employee Experience

Despite the need for companies to invest more in technology, there are constraints facing them including cost issues and lack of support to implement new technologies. With companies investing in technologies post pandemic, there are gaps that need attention for organizations to bounce back and leverage technology as digital transformation continues. For example, investing in workplace infrastructure is an area companies have experienced problems and need effective solutions.

The good news is that organizations are using employee experience as motivation to invest in technology and become data driven. The report by The State of Workplace Tech highlights this shift with companies focused on agile systems? and adjusting workspaces to boost employee productivity.

Detecting Depression from Speech

Voice recognition technology continues to evolve and startups are securing funding in the voice recognition space as the industry shows great promise. Detecting depression from speech is one area gaining popularity with Ellipsis Health, a startup using ML algorithms to understand the connection between the brain and nervous system.

Ellipsis Health reviews speeches through ML and deep learning and then estimates depression levels in a person. So far, startups in health care using AI including Kintsugi, Sonde Health and Vocalis Health are gaining traction with new venture funding for these startups. Ellipsis Health is using transfer learning? and deep learning algorithms to link thinking disorders and speech to depression which then gives an individual score.

Omnichannel Analytics and Conversation Intelligence

Customer intelligence is accelerating and companies understand the importance of customer engagements across multiple channels. Businesses struggle engaging with customers across email and social media but startups including Observe.AI are changing the experience by combining voice and text to enable smooth customer interactions.

Engaging with customers in a single dashboard is effective and facilitates smooth communication with customers. Observe.AI acquisition of ScopeAI attests to the growing demand for conversational intelligence? among enterprises. By combining text and sound under one dashboard, this capability from Observe.AI will assist companies optimize customer interactions by using insights in real time.

Specificity Constraint of AI

AI solutions tailored for different business use cases enable companies to understand their operations and make sense of data to drive business results. Despite the adoption of these AI solutions, the specificity problem still persists as AI solutions? released to the market have the assumption that all businesses experience similar problems.

However, the truth is that businesses have different dynamics which make them different and in need of specific technologies. This means one AI solution might work for one business and not make sense for the other. Specificity problems with AI adoption is the current roadblock facing many companies that feel that their tech stack does not generate value. This is where organizations should reevaluate their use cases and adopt AI solutions that address their pain points.?

Works Cited?

1Responsible AI Systems, 2Biased Algorithms, 3Sentiment Analysis, ?Ethical AI, ?Agile Systems,? ?Transfer Learning, ?Conversational Intelligence,? ?AI Solutions