The hope and the hype of AI
19th September 2024 | Journal Of Sales Transformation
Is AI ready to transform sales?
From AI-assisted Raybans that can initiate a call (look cool while you’re phoning clients from the beach) to email subject line generators (fill your prospect’s inbox with annoyingly zany emails), AI has been top of the hype charts ever since Chat GPT broke onto the scene almost two years ago (November 2022). Since then, a plethora of salesrelated applications have been pitched to an enthusiastic audience that understandably wants to boost numbers while making selling more efficient.
There is no doubt that AI is a genuinely hot topic. For example, Salesforce is currently talking about moving “from bots to agents”, whereby somewhat smart AI copilots that are good at straightforward work will be superseded by so-called autonomous AI or “agentic systems”. On 22 August, Salesforce introduced two new fully autonomous AI sales agents: Einstein Sales Development Rep (SDR) Agent and Einstein Sales Coach Agent, which will be generally available in October.
According to Salesforce, “AI copilots are almost like interns or new hires: super smart and good at straightforward work, but need guidance and oversight to perform at their best.” Autonomous AI, meanwhile, “can be thought of as trusted digital colleagues as opposed to assistants, and will provide game-changing advantages for organizations.”
AI sales applications
Of course, the Salesforce offerings are only the tip of the iceberg. AI can help save time by automating tasks, allow you to manage and prioritize leads better, and work to improve outreach and prospecting. For instance, there are multiple AI applications that can generate personalised emails and help you define your target audience, find leads, and accelerate online research about prospects. There are sales-related applications that can assist in building a personal brand on LinkedIn, create personalised outreach videos, summarise sales meeting recordings, plus multiple other applications.
To summarise, however, AI is largely focused on two aspects of the sales process: 1) bots that automate customer outreach and lead qualification, and 2) internally focused tools that enhance sales team efficiency through conversational interfaces for analytics and lead information. As such, AI tools help to free up sales and marketing teams to focus on more complex activities, so enabling them to establish stronger client relationships and drive business growth.
Potential downsides
That’s the theory, but what about the downside? We’ve probably all worried about the potential job losses that the AI revolution may bring – and this is already happening in some areas. In 2022, data from Socius revealed that 14% of workers had already experienced job displacement due to automation or AI, although the fear of job losses seemed to outweigh the actuality.1 But that was two years ago.
We’ve probably all seen the reports of GenAI making stuff up – so-called AI hallucinations.2 This is very much a danger (in the context of online disinformation) and dependent on how AI models are trained. It may be a particular issue for companies because certain AI models may be too generalised to reflect the nuances of individual company datasets.
According to Close CEO Steli Efti, “The issue with all the AI tools is that they’re based on specific datasets. They don’t include proprietary data that only you own. And because of this, AI won’t tell you who your customers really are, what they do, what patterns they share, and what’s going on in their heads.”3
Beyond this, there is the wider question of governance, something we explore in greater depth later is this special AI edition of the Journal (see “Walking the talk with AI” page 14, “Safe, ethical and compliant” page 16, “AI in sales: view from our experts” page 20, “A technology teammate” page 27 and “The power and pitfalls of AI” page 30) .
Environmental concerns
Since the rapid rise of generative AI, there have been growing concerns about the energy demands of the technology, especially large language models (LLMs). As AI has risen in popularity so too have its power-consumption demands, which put pressure on power grids and raise carbon emissions. According to one calculation, a single ChatGPT query consumes 15 times more energy than a Google search query. This is highly approximate and based on the estimated energy consumption of a Google search query being 0.0003 kWh per query and a ChatGPT-4 query using 0.001-0.01 kWh per query, depending on the model size and number of tokens processed.4 For comparison, it takes approximately 0.06 to 0.12kWh to boil a litre of water to 100°C in an electric kettle.5 Another source suggests that ChatGPT-3’s overall energy consumption equates to 1,248 MWh; Bing, which uses OpenAI ChatGPT-4, is calculated to use 7,200 MWh.6 Some other AI models are even more power hungry, while others are less so. Meanwhile Goldman Sachs reports that AI is set to drive an 160% increase in datacentre power demand.7 Whatever the precise figures, it is food for thought in the context of energy transition goals designed to tackle climate change as AI expands over the coming years.
Until recently, AI has been very much in the hype phase, yet it remains to be seen whether the technology will generate the financial returns anticipated by companies. For sure, chip manufacturer Nvidia has been riding high on demand for its advanced processors that are so essential to training AI models. However, according to Martin Peers at The Briefing, “The worry on Wall Street is that, if no one but Nvidia gets a return on AI, the bottom will eventually fall out of the AI market. That would be bad for Nvidia but also not great for big tech companies spending the equivalent of a small country’s GDP on capital expenditures right now to build new AI-powered data centers.”
Our view is that AI is here to stay and will usher in a period of significant transformation in business. Currently, the technology is poised to exit its hype phase. This edition of the Journal takes a detailed look at what lies beyond the spin around AI in sales, what traps we should avoid, whether AI is up to the sales job, the governance issues, and of course, what AI is currently good for.
1 Eric Dahlin et al, “Are Robots Really Stealing Our Jobs? Perception versus Experience,” Socius, 17 October 2022, https://journals.sagepub.com/doi/full/10.1177/23780231221131377
2 https://en.wikipedia.org/wiki/Hallucination_(artificial_intelligence)
3 Joanna Kaminska, “We Tested 13 AI Sales Tools: Here’s What You Need to Know About Them”, Close, 9 April 9 2024 https://www.close.com/blog/ai-sales-tools#:~:text=Our%20CEO%2C%20 Steli%2C%20says%20this,going%20on%20in%20their%20heads.”.
4 Reddit thread: https://www.reddit.com/r/aipromptprogramming/comments/1212kmm/according_to_chatgpt_a_ single_gpt_query_consumes/?rdt=63058.
5 “How Much Energy Does a Kettle Use?” A Tidy Mind, https://www.atidymind.co.uk/how-muchenergy- does-a-kettle-use/#:~:text=How%20Much%20Electricity%20Does%20a,would%20 require%20approximately%200.06%20kWh.
6 “AI Chatbots: Energy usage of 2023’s most popular chatbots (so far)”, TRG Datacenters, https:// www.trgdatacenters.com/resource/ai-chatbots-energy-usage-of-2023s-most-popular-chatbots-sofar/#:~: text=ChatGPT%3A%20Energy%20Consumption%20in%20MWh,and%20consists%20of%20 96%20layers.
7“AI is poised to drive 160% increase in data center power demand”, Goldman Sachs, 14 May 2024, https://www.goldmansachs.com/insights/articles/AI-poised-to-drive-160-increase-in-powerdemand.