humAIn Principles workshop
Last week Ben Cooper and I partnered with humAIn conference in Sydney to workshop and define AI principles, guidelines and guardrails with marketing and media professionals from across Australia.
humAIn is Australia's only annual AI conference for media, marketing, and the creative industries, and is put on by Unmade.
In the lead up to the conference we worked closely with Cat McGinn from humAIn to co-design an approach that could help us understand the level of ambition, adoption and challenge being faced by the industry. We agreed to combine an industry survey, with an in-person working session at the conference, all designed to generate enough insight to produce useful and insightful outputs that we could share with attendees on the day.
The survey responses highlighted the lack of AI maturity in industry, exposing a lack of organisational preparedness, with large gaps in AI familiarity, adoption and support and a broad mix of attitudes from apprehension around job losses to opportunities for growth and career development.
Marketing & media professionals overwhelmingly believe there’s not enough organisational or regulatory guidance on AI
Over 85% of respondents say they receive "Limited / no organisational guidance on AI"
Nearly 80% of respondents felt there was not enough regulation in place for AI
The expected benefits and challenges faced by the industry were clearly expressed by survey recipients
Increased efficiencies, cost reduction and better data-driven decisions are seen as the top benefits of Generative AI
While ethical considerations, job displacement, bias and discrimination are seen as the top concerns about Generative AI
“It’s here to stay, trust me so make it work for you, not instead of you.”
Leadership and generational gaps in optimism and adoption of generative AI
One interesting observation in the data was the difference between C-Suite respondents and others, showing a potential disconnect between reported levels of familiarity with GenAI tools and how optimistic they are about their application. These findings are perhaps a cautionary tale for leaders within media and marketing organisations to ensure they are keeping their teams informed about decisions, plans and benefits of AI adoption, as there’s a significant amount of scepticism around AI, particularly from those 34 and under.
Senior leaders also report higher levels of adoptions than others, who believe their organisations have more limited, exploratory or restricted adoption of GenAI.
Respondents also overwhelmingly foresee impacts on employment as generative AI matures within the marketing and media industry with over 60% expecting AI will replace workers.
Potential differences in opinion between media and marketing leaders and those within their businesses is also reflected in a significant gap between younger and older respondents, with those under 34 much less optimistic and much more concerned about ethical use of GenAI
In preparation for the workshop at humAIn, we also analysed survey respondents answers to more open ended questions to identify themes and principles we could take into the working group.
Drafting humAIn guidelines
More than just facilitating a discussion on what AI principles might mean for our industry, we wanted to move from defining principles to establishing guidelines and guardrails based on the working groups input.
We set ourselves a challenge to conduct the workshop at the beginning of the conference, and then play back our findings at the end of the day.
In order to achieve this ambitious deadline, we enlisted the help of AI to ensure we could go from unstructured inputs from the working group to structured outputs ready to share back to conference in under 5 hours.
Step 1: Capture workshop outputs
Participants had captured a number of actions and behaviours they hoped to see or to avoid in their organisations with either “We will…” or “We will not…” statements. We turned to ChatGPT 4o to help capture these using the new multi-model AI inputs and uploaded photos of the postits. ChatGPT did a great job of transcribing the outputs in seconds.
We now had 22 “We will…” statements and 18 “We will not…” statements
Step 2: Summarise the topics
“Now remove duplication and create a final list of 10 of each. These principles are for marketing and media industry.”
We then summarised the outputs, carefully reviewing the AI outputs against the inputs to ensure we were happy with the themes that were being captured. ChatGPT did a great job, the summarised themes captured what we wanted to say and reflected the conversation in the workshop well, but the language was pretty generic, bland and uninspiring. Definitely not what we would expect from humAIn.
Step 3: Making them memorable
“Brilliant. Please could you make these more irreverent and provocative to illicit action and memorability.”
Here are the AI Principles for the Marketing and Media Industry produced during the humAIn conference:
We will…
Unmask Everything: Be brutally transparent in all AI applications.
Put Humans First: Design AI to serve people, not the other way around.
Reward the Good Guys: Financially back companies that stick to their ethical guns.
Teach the Masses: Make AI knowledge accessible to everyone, everywhere.
Trust, But Verify: Scrutinize every AI output and data source.
Spill the Beans: Share all successes and failures openly to help us all improve.
Play by the Rules: Follow all privacy laws and regulations to the letter.
Give Credit Where It's Due: Acknowledge everyone who contributes to AI development.
School the Team: Continuously train teams to be AI-savvy.
Champion Ethical AI: Promote balanced, ethical use of AI tools and tech.
We will not…
Ditch Our Principles: Stick to our ethical guns, no matter what.
Chase Hype: Stay grounded and avoid the AI hype-train.
Worship Machines: Never let AI overshadow human importance.
Let Bias Slide: Actively hunt down and eliminate AI bias.
Use Dirty Data: Only use clean, legally trained AI models.
Replace Humans: Use AI to boost human roles, not replace them.
Compromise Privacy: Guard people's privacy and security fiercely.
Rip Off Ideas: Respect and protect others' intellectual property.
Create Noise: Avoid pumping out useless or misleading AI content.
Wait for the Law: Go beyond the law to ensure our AI ethics are top-notch.
We felt these guidelines and guardrails were a great reflection of the conversation we had facilitated in the morning, but Ben wasn’t done - he wanted some powerful statements to share back with attendees at the end of the day
Step 4: Create a rally-cry for the industry
“Great. Turn these into a pitch and rally cry for the industry”
We will…
We will unmask everything, bringing brutal transparency to all AI applications. We will always put humans first, designing AI to serve people, not replace them. We'll reward the good guys who stick to their ethical guns and teach the masses, making AI knowledge accessible to everyone. We'll trust but verify every AI output and data source, spill the beans by sharing our successes and failures openly, and play by the rules, adhering to all privacy laws. We'll give credit where it's due, acknowledging everyone who contributes, and continuously school our teams to be AI-savvy. Most importantly, we'll champion ethical AI, promoting balanced and responsible use of tools and tech.
And we will not…
We will not ditch our principles or chase the AI hype train. We refuse to worship machines over humans and will not let bias slide in our AI applications. We won't use dirty data from illegal sources or replace humans with AI. We'll fiercely guard people's privacy and security and will never rip off others' intellectual property. We will not create noise with useless or misleading content, nor will we wait for the law to enforce AI ethics. We pledge to always strive to exceed legal standards and set the bar higher.
“Shorter, sharper”
Rally Cry for the Marketing and Media Industry
We Will:
"We will unmask everything with brutal transparency. Put humans first. Reward ethical champions. Teach AI to everyone. Trust but verify. Share our wins and losses. Follow privacy laws. Credit contributors. School our teams. Champion ethical AI."
We Will Not:
"We will not ditch our principles. Chase AI hype. Worship machines. Allow bias. Use dirty data. Replace humans. Compromise privacy. Steal IP. Create noise. Wait for the law."
”Join us in leading ethical AI. Let's make tech serve humanity and set the standard for responsible innovation. Together, we can make AI work for everyone.”
In conclusion, I was delighted by the assistance we got from our AI partner in analysing survey data, summarising verbatim answers and helping us identify themes in the research, helping capture the output of the workshop, rapidly synthesising the guidelines and then drafting a rally-cry for the industry.
Not only did AI help us meet our ambitious deadline of turning around our workshop findings before the end of the day, our workflow was so efficient we did most of the hard work in the conference breaks, and were able to enjoy the rest of the outstanding conference presenters.
Our thanks to Cat McGinn, Tim Burrowes, Belinda Cusack and Gabrielle Stolojan and the rest of the humAIn team for supporting us with this crazy idea, and to all the humAIn working group attendees for making it such a fun and insightful day.
Contact us about our AI Strategy Sprint to help develop AI principles, responsible use policies and guidelines for your organisation