Canva’s CTO encourages all 5,000 employees to find the best AI tool for them

“We want to have 1,000 flowers,” says Humphries of this embrace of what he calls an “acceptable” license.

Certainly, Canva’s security and trust teams still vet all potential AI tools before employees are allowed to use them. And the company said it rejected vendors who did not provide clear assurances about data protection and sometimes cybersecurity practices.

But Canva’s overall intention is to send employees a strong message that they can play a major role in moving their course forward with AI. Humphreys says that if it acts as a custodian for all the technology systems it is deployed, it is a rethink of a more traditional approach. He says that now everyone is involved in research and development when it comes to AI.

“We want them to be inherently motivated to understand that their work has essentially changed,” says Humphries.

That also poses major challenges for Canva’s Humphreys and other C-sweet leaders. They assume the responsibility to educate the workforce on the best ways to use AI to increase productivity.

In July, Canva will be holding “AI Discovery Week.” This includes a three-day study course and a two-day “Hackathon” that gives employees considerable time to experiment with AI from regular work-day tasks. The entire organization is invited to attend, and the courses offered will be taught by Canva employees, including combinations of technologists and self-taught.

Every two weeks, Canva hosts an AI-centric virtual event that creates up to 2,000 participants. The company allows AI practitioners to share their favorite use cases. One employee shared how to create custom GPT, a specialized version of OpenAI’s GPT model, to perform certain actions, allowing managers to compare employee performance notes and career goals. This saved us time to draft the biennial performance management assessment at Canva.

In yet another example, engineers who have mastered AI coding tools such as Github Copilot, Cursor, and Windsurf are encouraged to not only effectively promote these tools, but also show how to display restrictions and show where human input is needed. Their output is always reviewed by the peer. Peers are also encouraged to use AI to understand their code.

Humphreys says new alumni have disadvantages when it comes to coding, as he knows how to use these tools, but doesn’t have the ground experience to determine whether AI writing code is good or bad. To that end, Canva is investing in internal mentoring with more senior employees and has created custom GPTs based on roughly 300 peer-reviewed articles written by Canva engineers. Junior employees can ask their fellow human peers for GPT questions.

“I think it’s a challenge for the industry and a way to get junior engineers to make that transition into seniors,” Humphries said.

Humphreys says the “majority” of the company’s engineers are experimenting with AI, with 50% using these tools every day. He hopes that the number will rise to 80% by the end of 2025. Software engineers are about 30% productivity, comparable to what many other CTOs and chief information officers see, but Humphrey says he doesn’t stick too hard about it. He’s really focused on adoption.

He joined Canva as a senior engineer in 2014 and rose several times before being promoted to the role of CTO in September. Humphreys oversees all technologies, including IT, product development, infrastructure, security, and data science. Externally, the recent launch includes AI-powered collaboration spreadsheet products powered by Google’s VEO 3 model, AI video generation with audio, and AI video generation. The company says Canva AI now uses synchronous audio to generate 8-second video slips.

For AI’s internal applications, Humphreys emphasizes the importance of keeping humans in a loop, whether AI is being used to source code, write press releases, prepare for job assessments for colleagues, or develop marketing plans. Employees are always reviewed for all AI Assist jobs.

“Humans are ultimately owners,” says Humphries. “They really have to own that output.”

John’s needs

Send your thoughts and suggestions to CIO Intelligence here.

News Packet

Meta employs top AI talent, but skepticism about strategy is swirling. Mark Zuckerberg spends a lot of money on AI, including AI A $14.3 billion deal to invest in scale AI and hire co-founder Alexandr Wang,search Acquisition of a company Efforts like a video startup runway AI Bring it in Daniel Gross, CEO and Founder of Safe Superintelligence, and former Github CEO Nat Friedman. But as Bloomberg NoteMeta Insider believes its leadership is worth noting in Meta’s AI division, with external applications disappointed and the departure of most employees who built the Meta’s leading language model, Lama. “The Lama 4 announced this year was an absolute failure,” DA Davidson analyst Gil Luria told CNBC on Friday. “Meta cannot afford to fail to have a major AI model. So they are now in the market where they are desperately trying to trade AI teams.”

AFLAC is the third insurance company to hit this month’s cyber attack. AFLAC, which provides home and life insurance to over 50 million policyholders; Disclosure Last week, cybercriminals could have violated the company’s US network and may have accessed customer personal data, including social security numbers. Aflac says he identified a bad actor in the US network on June 12th and believes he stopped the intrusion within hours. The company also believes the attack is part of a broader campaign against the insurance industry, with two other insurers (Erie Insurance and Philadelphia insurance companies) also disclosed the violations in June. Beyond insurance, cyber is also a growing fear for federal officials I warned The attack could grow amid escalating tensions between the US and Iran.

Recruiters are buried beneath a mountain of growing AI-enabled recruitment applications. In particular, following Amazon CEO Andy Jassy’s, there is growing concern over the impact of AI on white-collar workers. comment About the company’s future personnel needs and Sam Altman I’m saying AI can rival a PhD in several tasks. But what about the actual impact of AI on employment practices? New York Times Deep Dive reports that the number of applications submitted on LinkedIn has increased by 45% over the past year. The platform averages 11,000 applications per minute, and generative AI tools contribute to this massive amount of content, making it difficult for recruiters to decide which resume, or even the applicants themselves.

The former Openai CTO AI startup has raised $2 billion. Thinking Machines Labs, an AI startup founded in February by former Openai CTO Mira Murati, has officially closed in the seed round. value According to the company, the company has $10 billion. Financial Times. Beyond the vague language on websites, there is little known about thinking machines, including “a wider understanding of AI systems, customizable and generally capable.” The company is like that It’s packed A big funding round vault considering a growing machine with AI talent from Openai and a group of over two dozen US AI startups that already have It was raised Over $100 million in 2025.

Recruitment curve

Data concerns are the biggest reason that most AI projects aren’t on time. a investigation Of the 1,000 business decision makers, 69% of organizations found that at least one of the planned AI initiatives reported being behind schedule. The top three main causes of these AI project delays were concerns regarding data quality and availability (39%), challenges for integration with legacy systems (37%), and regulatory uncertainty and compliance challenges (36%).

The findings conducted by Zogby Analytics in May and supported by AI management companies found that AI proves AI, with 86% of companies playing an active role in training and fine-tuning custom AI models, while 55% say the process is more difficult than expected.

The majority of organizations that voted (81%) said they had an annual AI budget of at least $1 million, with one in four companies investing more than $10 million.

Job Radar

employment:

– Candid health teeth Looking for a major software engineerbased in New York City. Posted salary range: $238K-$322K/year.

– zip co teeth Looking for a Vice President of Engineering in the USbased in New York City. Submitted Salary Range: $240k-$30K/year.

– ipicatal teeth Looking for a technical product manager, SVPbased in New York City. Posted pay range: $200k-$240K/year.

– Grant Thornton teeth Looking for the AI ​​headbased in New York City. Posted salary range: $206.3K-$395.4K/year.

Hired:

– McDonald’s (Fortune 500 No. 165) Promotion Valerie Ashbaugh served as US CIO after spending more than four years leading the restaurant chain’s global commercial products and platform team. Before joining McDonald’s in 2021, Ashbo spent more than eight years in various leadership roles at Nike, including SVPs in Product, Planning and Manufacturing. She previously worked for Boeing, Dell and Accenture.

Every Friday morning, the weekly Fortune 500 Power Moves column tracks the Fortune 500 Companies C-Suite Shifts–See the latest edition.

– Legend It has been named As CTO, Jim Scholefield oversees the technology strategy, innovation and execution of live event companies across all digital platforms and systems. Scholefield joined Marriott hotel chain Legends, where he served as Chief Information and Digital Officer. He also Merck & Co. , played leadership roles at Nike and Coca-Cola.

– Love’s Travel Shop announcement Tim Langley-Hawthorne has been appointed CTO and will soon be effective, where he will drive product development and AI deployment. Before joining Truck Stop Operator, Langley Hawthorne served as CIO for car rental company Hertz and served as executive technology leadership roles for Hitachi Vantara, Western Union and Information Services Group.

– Massachusetts Institute of Technology Appointment Gaspare Loduca is an effector on August 18th as the CIO and VP of Information Systems and Technology. Roduka joined MIT after serving as VP of IT at Piercolumbia University. At MIT, Loduca oversees IT infrastructure and services, supporting research and academic companies across student and management systems, cloud services, cybersecurity, and customer support.

– Provider It has been named With Qingqing Ouyang as Chief Product and Technology Officer, he leads the investment in digital innovation from online companies to business markets connecting alcohol suppliers, distributors and retailers. Most recently, Ouyang served as CPTO at real estate software company Ojo Labs, where he served as leadership roles at Tech Giants Dell and Oracle.

– Denver Broncos I have it I was hired Daniel Brusilovsky joined the NFL team as CTO after serving as vice president of technology for the NBA’s Golden State Warriors. Brusilovsky oversees all technology services and systems for teams at the team’s training facilities Bronco Spark and Mile High Stadium Empower Field.

– floqast announcement He appointed Sri Ramalingam as CTO, joined the accounting software provider, and advanced AI-powered automation features, including the recently launched Floqast AI agent. Ramalingam recently served as engineering SVP for software developer Harness. He previously served as Vice President of Engineering at Zoom and held leadership roles at Cisco and Plantronics.