Transitioning, according to Gartner, less than 25 percent of government organizations will have citizen-facing services powered by AI by 2027. Governments are hesitant due to concerns about empathy in service delivery and meeting community expectations. Risk and uncertainty, as well as a lack of traditional controls, are slowing down the adoption of AI at scale.
HP’s Work Relationship Index reveals that Artificial Intelligence (AI) can potentially improve relationships with work. The study shows a strain in the relationship between individuals and work, but AI is seen as a key to unlocking better relationships. Business leaders must educate employees on AI’s potential and guide its integration to foster understanding and acceptance. The survey also highlights the need for upskilling initiatives to empower the workforce to embrace AI technologies.
Four studies:
A recent study from the Technical University of Denmark suggests that ChatGPT exhibits significant gender biases, associating women with certain professions and activities while linking men to others.
An experiment conducted by Bloomberg found evidence of gender and racial bias in OpenAI’s GPT-3.5 and GPT-4 models when ranking resumes for different job descriptions. The study revealed unequal treatment and adverse impact on specific demographic groups, highlighting concerns about discrimination in AI-powered hiring systems.
AI models tested on election-related queries showed high rates of inaccuracy, harmfulness, and incompleteness, according to expert raters. Google’s Gemini, Mistral’s Mixtral, and Meta’s Llama 2 had the highest inaccuracy rates, while OpenAI’s GPT-4 was the most reliable.
IBM’s use of Adobe’s AI tools in marketing campaigns has resulted in productivity improvements, with a 10-fold increase in designer productivity. The tools, which generate images from text-based prompts, have allowed IBM designers to generate ideas quickly and create variants for different parts of campaigns.
Why do we care?
Despite my enthusiasm for AI, the data says we have a ways to go.
Organizations should foster an AI-positive culture that emphasizes AI as a tool for enhancing productivity, creativity, and job satisfaction. Upskilling and reskilling programs are essential to prepare the workforce for AI integration, ensuring employees feel empowered rather than threatened by AI technologies. There’s work to be done on addressing bias, and keeping humans in the loops, which we’ll get to more in a moment.