I’ve been involved in two discussions recently which have collided, for me at least, in a mix of hope and expectation.
Discussion#1 goes something like this.
Is the care economy a thing? And if it is a thing, is it a thing whose defining characteristic, fairly or not, is its stubborn resistance to many of the traditional tools and methods of productivity improvement? Valuable, necessary even, but recalcitrant and resistant to (some forms of) discipline?
Discussion#2 goes along these lines.
Artificial intelligence (AI) is roiling the settled rhythms and routines of pretty much every organisation and every sector it touches – which is everyone and everything – but is struggling to find its sweet spot in the human services sector. The care economy, apparently, is a problem child. Again. If you’re in the care business, whose “product” is quintessentially fashioned from deep connections one human to another, what really does the fun and fury of AI have to offer? Is it really more of a distraction with marginal benefits and significant costs (imagined and real)?
Both discussions – the care economy’s tussle with productivity and AI in the human services sector – are vital but muddled and frustrating in equal measure. And both are profoundly connected to each other in a bigger story about the role, purpose and value of the care economy in modern economies and in society.
I’ve written recently about the first discussion . With others, I’m still exploring some of the fundamental misconceptions and mistakes that tend to skew our attempts to think clearly about the care economy’s role and value.
Designed by care experts, not tech experts
My tracking of the second discussion introduced me to the work of Beam and an application they have pioneered – Magic Notes (powered by Beam) – which shows what can happen when AI comes up with ideas designed with and by care experts, not solutions designed by tech experts. It’s quite an important difference.
The work sits right at the intersection of productivity and AI.. Beam’s tagline hides the mission in plain sight: “we build AI tools that transform frontline productivity.”
They’ve built and grown a solution that speaks directly to the productivity debate in the care economy. They have shown how to fashion new tools and platforms that do the apparently impossible – to lift productivity and effectiveness across the care economy safely and ethically without jeopardising compassion, dignity, respect and effectiveness. To the contrary…
Beam was founded in 2017 to support prison leavers, people experiencing homelessness and refugee communities in the UK. Confronted with the operational and compliance pressures facing frontline teams, Beam built its own technology to enable higher-quality, audit-ready support at scale.
The result is Magic Notes (powered by Beam), an AI-powered application that captures conversations, meetings and other interactions between frontline workers and clients to turn then rapidly, safely and accurately into template reports and summaries whose dimensions have been fashioned from the ground up for each organisational and operational context given different policy and practice requirements in different policy domains and jurisdictions.
Bean is now working with over 350 organisations and Magic Notes is supporting over 80,000 frontline staff across government, not-for-profit and private sectors. It is working in live deployments in 6 countries. Including the UK, US and Australia.
Beam in Australia
Beam.au is now headquartered in Melbourne. It’s building a local team and long-term presence, including the resources for rigorous and sustained onboarding and training, to integrate Magic Notes into to the Australian human services sector. That’s already started with Life Without Barriers where an early prototype is now being rolled out across service teams.
“Beam are bringing a social purpose mindset to how we can augment the best of AI, so our people can invest in the crucial trust and relationships we have with communities.” Claire Robbs, CEO Life without Barriers
One of the biggest tests for the spread of AI tools and capabilities is human services, where high and rising opportunity and gathering demand is too often misaligned with tools and platforms struggling to fit the contours and rhythms of the “human” part of human services. In other words, the bit that matters.
Magic Notes is designed from the insights and experience of front line staff and people with lived experience, delivering big savings of money and time and witnessing step changes in productivity, quality and impact.
Making the right kind of difference
An independent evaluation by Warwick University in the UK tracked 8.5 hours a week saved on average by frontline workers.
On the safety and privacy side of the ledger:
- Frontline staff and clients confirm the accuracy of the conversation capture (meeting notes, recommendations for action, follow up responsibilities)
- Login is via Single Sign On and Two Factor Authentication
- The data collected stays private and isn’t used to train either Magic Notes itself or other applications
- Data is hosted on government-grade servers in Australia, not offshore.
The benefits ripple across every dimension of the work – lifting the quality of records and notes for compliance and regulation, building trust and engagement with service users and clients, workforce confidence and capability (with big payoffs for recruitment and retention, with up to 1 in 9 frontline workers leaving the sector after a few years) ) and not to mention innovation and organisational performance
I suspect one of the reasons for its success is the pivotal challenge it’s picked.
How can you capture the conversations and human interactions (meetings, conversations, interviews, therapy sessions, service reviews) which are so basic to everything the sector does in a way that is accurate, safe and ethical and allows energy and limited time to grow relationships of care, support, dignity and respect? That’s the persistent imperative for any human services agency or service delivery organisation. It’s exactly the right point to apply the pressure of invention and “good tech for good”.
A big opportunity
There are few policy challenges more urgent or more complex than lifting the purpose, significance and performance of the care sector at the heart of our economy and society.
Good people with guts, determination and expertise are tackling that task head on with an agenda of inspiring innovation and pragmatic modernisation. But it often feels that we’re simply not turning palpable progress into the momentum we need for requisite impact with all necessary speed, intensity and scale.
Widespread adoption of the capabilities and, more importantly, the culture and care-driven, custom-designed and grounded tools like Beam’s Magic Notes strikes me as exactly the kind of momentum we need towards a care economy that integrates lifting productivity, privileging respect and relationships and intelligently embracing the best of technology’s powerful potential.
Leave a Reply
Facts Only
* Discussion #1 asks if the care economy is a thing and if it resists traditional productivity improvement methods.
* Discussion #2 questions AI's role in the human services sector, specifically concerning the care economy.
* Beam developed Magic Notes, an AI application using Beam technology.
* Magic Notes captures conversations, meetings, and interactions between frontline workers and clients to create template reports and summaries.
* Beam was founded in 2017 to support prison leavers, people experiencing homelessness, and refugee communities in the UK.
* Magic Notes is used by over 350 organizations and supports over 80,000 frontline staff across government, not-for-profit, and private sectors.
* The application is deployed in live settings in six countries, including the UK, US, and Australia.
* A Warwick University evaluation tracked an average saving of 8.5 hours per week for frontline workers.
* Data capture includes meeting notes, recommendations for action, and follow-up responsibilities.
* The system uses Single Sign-On and Two Factor Authentication.
* Collected data is not used to train Magic Notes or other applications.
* Data is hosted on government-grade servers in Australia.
Executive Summary
The core tension explored is the role of the care economy in relation to productivity improvements and the impact of Artificial Intelligence (AI) within the human services sector. One discussion questions whether the care economy, defined by deep human connections, resists traditional productivity methods. The second discussion explores whether AI offers meaningful value in the human services sector, specifically challenging whether it can offer substantial benefits beyond distraction given the sector's reliance on interpersonal relationships. This tension is framed around the idea that AI tools must be developed by care experts to ensure they align with the core values of compassion, dignity, and respect, rather than being purely technology-driven solutions.
The example provided is Beam's Magic Notes application, an AI tool designed with input from care experts to capture frontline interactions into standardized reports. This tool is presented as a mechanism for increasing productivity and effectiveness in the care economy while attempting to safeguard relational aspects. Empirical evidence cited includes an evaluation by Warwick University showing 8.5 hours per week saved by frontline workers. Safety measures are detailed, including single sign-on authentication, data privacy, and hosting on government-grade servers in Australia. The narrative concludes that the successful application of such tools is dependent on addressing the fundamental imperative of how to capture human interactions safely and ethically within service delivery organizations.
Full Take
The narrative constructs a compelling argument for an intersectional solution, positioning the tension between productivity demands and relational care as a central problem needing innovation. The success of Beam's work pivots not just on the technical capability of AI, but on embedding the constraints of the human services context—dignity, trust, and relationship-building—into the design from the outset. This suggests that technological adoption in sensitive sectors is less about optimizing existing workflows and more about finding ways to capture essential relational data in a manner that preserves the quality of human interaction.
The pattern observed is an attempt to neutralize the perceived threat of technology by anchoring its implementation to an unshakeable, non-negotiable ethical mandate derived from frontline experience. This counters a potential trajectory where AI is deployed merely as a cost-saving mechanism, which often leads to superficial capture of data that erodes trust and effectiveness. The focus on building tools "designed by care experts, not tech experts" functions as a corrective against the dominant pattern in technology adoption—the imposition of external, abstract solutions onto complex human realities.
The implications point toward a systemic need for valuing relational infrastructure alongside operational efficiency. If AI can safely augment frontline productivity without compromising compassion and dignity, the opportunity lies in using this augmentation to create more space for the essential work of care. The challenge shifts from *if* AI can be useful to *how* it can be responsibly molded to serve the inherent purpose of human services rather than merely optimizing them. The lingering uncertainty centers on whether this success is scalable enough to shift the broader momentum required for comprehensive societal change in the care sector, or if it remains an impressive localized pilot against a massive structural inertia.
Bridge Questions: If ethical design is the key differentiator, what formal frameworks are necessary to ensure that "care expert" input is universally prioritized over pure technical feasibility in all AI development? How can organizations move from successful pilot deployments to establish the necessary cultural momentum for widespread adoption of these relationship-centered technologies? What are the long-term social and economic consequences if productivity gains in care sectors fail to translate into sustained improvements in human dignity and workforce retention?